Product Watch: Compaq Sets Standard For Flexibility. IBM To Sell Supercomputing Sofware. MontaVista Unveils Linux Kernel Prototype.

September 15, 2000



Houston, TEXAS — Compaq Computer Corporation introduces a new PCI RAID controller architecture that delivers unprecedented flexibility through a modular design. The Smart Array 5300 (SA-5300) controller series meets customers’ needs for rapid data growth, better server performance, robust fault tolerance, and a future connection to the storage network. “Compaq will soon sell its two-millionth Smart Array controller, evidence of the confidence customers place in the most trusted name in PCI RAID controllers,” said Ron Noblett, Vice President, Server Storage, Compaq Computer Corporation. “Compaq’s innovation continues now with the SA-5300, offering customers an architecture that enables growth.”

With the addition of a storage area network (SAN) access module available next year, the SA-5300 will create a platform for bridging between the direct connect storage and Compaq’s SAN infrastructure. Once connected, customers may take advantage of enterprise-wide back-up solutions, reducing the cost of managing their direct connect server storage, and protecting their current investment. The SA-5300’s new design allows buyers to increase cache, increase SCSI channels, and add Advanced Data Guarding (RAID ADG), a new RAID technology. Customers will find common software tools that help simplify storage management by offering consistency across the product line and from generation to generation. Offering data compatibility among all Smart Array controller models allows quick and easy upgrades to enhance data performance, capacity and availability, and customers can easily deploy and re-deploy drives with data intact. Information about Compaq and its products and services are available at .


New York, N.Y. — IBM will begin selling supercomputing software to help big Internet companies better manage burgeoning data demands, Big Blue said today. Until now, the software had been used to let one workstation manage supercomputers made up of multiple refrigerator-size units. IBM will now sell this single-control data-management programming to operate clusters of its RS/6000 S80 Web server computers. Before this, Web servers could be linked but lacked software to let them be managed easily from one workstation.

Charles Schwab, for example, uses an IBM supercomputer to process transactions. Although most companies don’t require that level of computing power, they may need more than one Web server and a way to link them in order to gain processing power as their businesses grow. The base system of two linked RS/6000 S80 servers with supercomputing software costs $705,000, IBM said.


San Diego, CALIF. — MontaVista Software has taken the wraps off what it calls a “hard real-time fully preemptable Linux kernel prototype,” based on Linux kernel 2.4. The company claims its kernel modification has the power to revolutionize the application responsiveness of Linux, while fully preserving the Linux programming model. The prototype kernel modification is currently showing better than a 10-fold improvement in worst-case kernel responsiveness, but is expected to reach gains of 100-fold or better, by the time the technology is incorporated into MontaVista’s Hard Hat Linux distribution in the January 2001 timeframe. Kevin Morgan, MontaVista’s Vice President of Engineering, says “the kernel modification results in a “relatively fully preemptable kernel at this point in time. The long locks that are held in a SMP environment are being implemented as a no-preempt region, which we feel is a good starting point for this approach. Our goal is to eventually get those to much shorter periods. We’ve got a lot of measuring work and tuning work to do, as well as refinement of the specific techniques to both drive the responsiveness to even better numbers and to improve some of the architectural attributes. Right now what we’re putting out is a prototype, in the best spirit of ‘release early, release often’.”

Morgan points out that the enhanced kernel responsiveness is not merely useful for esoteric real-time industrial process control systems, but rather is of value even to today’s desktop PC users. “This technology has use in all segments of the market, including the desktop, for the simple reason that desktops these days require multi-programming and multi-processing, with live streaming media in action,” says Morgan. “For example, as a desktop user I want to be able to watch a movie and hear the sound, while also running a browser and my mail program. And when I use the mail program and the browser, I don’t want any glitches in the movie or sound. That really requires improvements in Linux responsiveness, and that kind of behavior really is a real-time problem, because a human is really pretty good at detecting glitches in continuous real-time operation. So, this technology is applicable on the desktop. And, of course, it’s highly applicable in the more traditional embedded control environments.” An obvious question, given the potential benefit to virtually all Linux applications – from desktop, to server, to embedded – is: why not add this capability to the standard, mainstream Linux kernel? In fact, that is precisely MontaVista’s intention. “MontaVista would like to see this technology, or similar technology, utilized as a foundational feature of Linux 2.5,” explains Morgan. “We are excited to offer this in prototype form, and look forward to receiving feedback from the all areas of the Linux community, as we continue its tuning and refinement.”

According to Morgan, a group of Linux game programmers recently petitioned Linus (Torvalds) to add specific preemption points in the 2.4 kernel, primarily for video throughput improvement. Their proposal was rejected, according to Morgan, due to the specificity of the fixed preemption points. “The kind of fully preemptable kernel technology that we are developing would definitely meet their needs,” says Morgan. “In the course of the dialog over adding specific fixed kernel preemption points, Linus indicated that he is leaning towards a fully preemptable kernel approach in Linux 2.5. We hope our prototype will help to spur and support that change, and possibly become the basis of that technology.” MontaVista’s Linux kernel enhancements also include a real-time scheduler which the company says provides extremely high performance and deterministic real-time application selection and dispatching, without altering standard Linux scheduling. MontaVista also offers an optional interrupt accelerator – in the form of RTLinux – which can be used to improve baseline interrupt responsiveness by better than 5-fold. The result of all these is what MontaVista terms a “closing of the gap between the real-time capabilities of Linux and proprietary RTOSes.” Significantly, these technologies are of fundamental value to mainstream Linux-based desktop and server systems, where multimedia and quality-of-service (QoS) demands are increasingly demanding high performance real-time responsiveness.


Loveland, COLO. — Terra Soft Solutions, Inc., a leading developer of Linux solutions for PowerPC microprocessors, has announced the offering of a portable, affordable, 8-node PowerPC cluster with Black Lab Linux installed and configured for parallel processing and development. Parallel processing refers to speeding-up code execution by dividing the program into multiple fragments that can execute simultaneously across multiple computers. “Not everyone requires a room full of nodes to benefit from high-performance computing. We recognized the need for a powerful, light-weight, Black Lab Linux cluster that could literally be wheeled from room to room in a laboratory, school, or code development environment. Parallel code development can be a bit tricky and it is often comforting to have the entire cluster next to your desk. Our clusters will help developers get their feet wet with PowerPC Linux parallel computing,” stated Kai Staats, CEO of Terra Soft Solutions.

Terra Soft’s first customer is MIICRO, Inc., an imaging technology company with a patented neuroimaging method for the evaluation of CNS drugs using positron emission tomography (PET). MIICRO will use the parallel system to develop its neuroinformatics services for the rapid screening of CNS compounds, and the development of high-speed image analysis tools. “Working with Terra Soft provides a great opportunity for us. Linux continues to play a vital role in the development of our commercial services and intellectual property,” states Declan Cooper, VP of Technology for MIICRO, Inc. “Their cluster makes possible for us to port our code from a uni-processing mode to parallel processing mode for considerably less than we had anticipated. Our research has shown the PowerPC to be a strong and fast platform, it is therefore natural for us stay with the same platform for parallel code development,” says Cooper.


London, ENGLAND — UK tape drive and tape library manufacturer M4 Data has expanded its MagFile tape library product range with LTO (Linear Tape Open) technology. LTO is a new tape standard agreed by IBM, Hewlett-Packard and Seagate, providing 100Gb of uncompressed data at 10-20Mb/s on a single 4″ square cartridge. Availability and pricing of the new LTO drive was not revealed.


Sunnyvale, CALIF. — MontaVista Software, Inc., developer of the Hard Hat Linux operating system for embedded applications, announces the immediate availability of a hard real-time fully preemptable Linux kernel based on current 2.4 software. This kernel revolutionizes the application responsiveness of the Linux kernel, while fully preserving the Linux programming model. The MontaVista hard real-time preemptable kernel provides a 30-fold improvement in application responsiveness of a standard Linux operating environment. MontaVista is targeting a January 2001 release of a standard Hard Hat Linux product based on this technology. “The MontaVista real-time strategy is to drive native real-time improvements in Linux,” explained James Ready, president and CEO of MontaVista software and RTOS industry pioneer. “Real-time capability is the final barrier to comprehensive adoption of Linux throughout the embedded systems industry. MontaVista’s hard real-time fully preemptable kernel technology advances Linux to the responsiveness attributes of proprietary kernel products. It does so while still providing all the benefits of open source, including rapid development and deployment based on a royalty-free model.”

A fully preemptive kernel is essential for highly responsive real-time systems. The immediately available preemptable kernel improves real-time application response from hundreds of milliseconds (worst case) to about twelve milliseconds. “We expect to achieve sub-millisecond (hundreds of microseconds) worst case application responsiveness by the time of general product availability in January,” says Kevin Morgan, vice president of Engineering for MontaVista. “Our design leverages the Symmetric Multi-Processing (SMP) locking facilities; this means that the design for real-time and the design for multi-processing in Linux are completely aligned. Our solution is an example of our 100% pure Linux product philosophy in action.” “MontaVista would like to see this technology or similar technology utilized as a foundational feature of Linux 2.5,” says Morgan. “We are excited to offer this in prototype form, and look forward to feedback from the Linux, embedded and real-time communities as we continue its tuning and refinement.”


Irvine, CALIF. — Vision Solutions Inc. (“Vision”), a global leader in the development and marketing of clustering, high availability, managed availability, logical partitioning and data replication, announced the general release of its award-winning Object Mirroring System, OMS/400 R7.0. This release delivers a significant increase in functionality, specifically related to clustering, remote journaling, MQ series and single system mirroring support. Additionally, improvements have been made to the release’s overall performance, ease of use and speed of installation.

OMS/400 is the most advanced data mirroring high availability product in the market today. In addition to its ability to synchronize enterprise-wide data, by simulcasting changes from a source system to more than 9000 target destinations, OMS/400 incorporates features such as self-healing techniques and tuning tools, built-in object refresh capabilities, no file/object limitations and mirroring. Vision Solutions’ Jose Bliek, executive VP product management, commented: “The release of OMS/400 7.0 is one of the most significant in the product’s long history. It is feature-rich with support for remote journaling and MQ Series, as well as full clustering capabilities. “OMS/400 7.0 has already been widely applauded by Vision’s beta-testers and we are highly confident that existing and new customers will quickly benefit from the improved performance and enhanced capabilities that it brings.”


Lebanon, N.H. — Fluent Incorporated, producer of Icepak thermal modeling software, announces version 2.0 of its Pro/Engineer-Icepak Interface. This interface continues to build on its successful integration of CAD software with thermal modeling – a logical next step in electronics design processes. “Virtual prototyping of a designs’ thermal, electromagnetic and mechanical performance has proven to be an effective short-cut in the race to market for Icepak users,” says Fluent’s Icepak Product Manager Rajesh Nair. “The new interface reflects our users’ input to bring these capabilities to a higher level with greater speed and efficiency,” he adds.

Specific enhancements in v2.0 of the Pro/Engineer-Icepak direct CAD interface include greater flexibility in object handling, eliminating repetitive functions and greater flexibility in working with shapes such as prisms and inclined plates. The pressures of the market are driving the demand for more and more sophisticated computer modeling tools. With this demand for faster and more accurate simulation tools has come a parallel need for these tools to work more effectively together. File format compatibility is not difficult to achieve, but the Pro/Engineer-Icepak interface’s level of integration is a higher challenge. “It’s one thing to have one program read another program’s files,” says Icepak Business Unit Leader Prabhu Sathyamurthy. “But to have a direct interface for your program within another program requires a high degree of trust and dedication to the user’s needs. Together with our Icepak development partners, ICEM-CFD Engineering, we focused on the users’ needs to achieve greater design integration. Typical CAD files are too information-dense to be effective for thermal analysis. Importing the data directly requires time-consuming cleanup and alteration in Icepak. With this interface Pro/Engineer’s powerful tools enable the user to prepare the file before importation to Icepak. This way users save time, with better results, and maximum design integration efficiency.”


Ann Arbor, MICH. — Distance Engineering announced the release of CADInSite training CDs for Solid Edge customers. This 2-disk set is titled “CADInSite for Solid Edge: Part Modeling Fundamentals.” “How often do you leave a class feeling that you understand things only to find out the next day that you have forgotten much of the material? CADInSite keeps the instructor at your side.” says Peter H. Smith, CEO, Distance Engineering. “This CD set has 76 narrated screen video lessons by the prolific modeling expert Walt Silva, also known as Doctor Walt. We wanted to bring to CAD users the confidence that comes from sitting next to a friend who has `done it all’ and is willing to show you the tricks of the trade.”

This new Solid Edge training program, narrated by Walt Silva, author of 11 CAD books sells for $84.95 (U.S.) and includes a set of 2 CDs.


Boston, MASS. — At the XML World Conference in Boston, KaanBaan Technologies, an emerging enabling technology company, released its 1.0 beta version of the XML-Broker, an advanced XML transaction server for e-business integration. KaanBaan has identified a major gap in the e-business transactions space and has filled this gap with the release of the XML-Broker. Companies seeking to build and implement XML transactional-based solutions are forced to patch together solutions using a variety of products from multiple vendors. This creates problems with configuration, integration and maintenance. Moreover, XML application programming can be crude, tedious and prone to programmer error, so it creates many ongoing maintenance issues.

The XML-Broker, an XML transaction server based on 100% Java J2EE technology offers a streamlined solution to the problem. The XML-Broker consists of a stand-alone transaction server, which is Enterprise JavaBean (EJB) compliant, and an application development environment for quickly developing XML components. The XML-Broker transaction server adheres to the Enterprise JavaBean 2.0 specification providing standard transaction management, resource sharing and an EJB compliant container. It also provides security using open standards including SSL and X.509 digital certificates. Other features include synchronous/asynchronous transactions, fault tolerance, dynamic load balancing, component caching and web based administration. Additional information about KaanBaan is available at .


Amherst, N.Y. — ATTO Technology, Inc., a worldwide manufacturer of SCSI and Fibre Channel storage connectivity solutions, announced the debut of the ATTO FibreBridge, 2200R/D Fibre Channel-to-SCSI bridge. The 2200R/D incorporates high-end features previously unavailable to smaller scale operations at an attractive price point. This product illustrates ATTO’s commitment to bridging the gap between mid-range and advanced storage technologies by offering the advantages of Fibre Channel-based connectivity to a broad range of companies.

The multi-platform FibreBridge 2200R/D is configured in a versatile enclosure that allows it to function both as a desktop bridge and a 1U-high, rack-mounted or drop-in device. The FibreBridge 2200R/D, which delivers up to 98 MB/sec. of sustained throughput, is aimed specifically at mid-range Storage Area Networking (SAN) applications. “The markets that are reaping the benefits of Storage Area Networking are growing at an astounding rate,” states ATTO President Timothy Klein. “These markets share a critical need for solutions that bridge the barriers between cost, performance and versatility. The FibreBridge 2200R/D is specifically designed to speed the transition to Fibre Channel-based storage by delivering features and performance commonly associated with leading-edge devices, at an attractive cost. Combined with design features that provide for easy expandability, the FibreBridge 2200R/D will undoubtedly fulfill a unique and growing market need.”


Minneapolis, MINN. — Ciprico Inc. announced the first NETarray RAID systems began shipping to visual computing customers worldwide. NETarray 1000 is the newest member of Ciprico’s Fibre Channel Family of Digital Storage Systems. It represents a leap forward in the development of high-performance RAID storage combining the best technologies of storage solutions for visual computing environments with those normally found in the enterprise computing world. “All of our markets will benefit from NETarray because in a single scaleable platform, we provide our customers both high-bandwidth and transactional storage,” said Ciprico President and CEO Robert H. Kill. “NETarray solidly brings Ciprico’s service and performance into applications and markets other than our traditional high bandwidth environments. It builds on our base of products and services, squarely meeting the needs of today’s environments requiring high I/O applications such as internet streaming and video editing.”

Storage capacity and application needs keep growing, but Information Technology management is expensive and scarce. NETarray recognizes the need for storage to be managed like other IT resources. The NETarray is the first storage solution to provide the option of seamless out of band management, freeing up processing and increasing I/O performance. Storage can be easily managed by linking to the Ethernet through any web browser to tools already used for other IT resources. Whether used for broadcasting, streaming video over the internet or running an application within government, NETarray’s out of band management revolutionizes storage management by keeping it simple. NETarray is designed with high storage density. Each NETarray enclosure supports one or two high-performance hardware RAID controllers and can be configured with any combination of half-height Fibre Channel Arbitrated Loop (FC-AL) disk drives, from a minimum of four, to a maximum of ten. A single rack scales to holding up to 100 drives in less than a 30U rack space, totaling 7.3 TB per rack, using 73 GB drives. In a small footprint, NETarray delivers screaming I/O performance, and provides open scaleable storage for growing video streaming, Internet and other I/O intensive environments.


London, UK — LaCie announced it will unveil its new hard drive boasting a rotational speed of 15,000 rpm at Apple Expo 2000. This external hard disk is equipped with the new Ultra 160 interface and will be available in the popular 18GB capacity. Ultra160 SCSI LVD is a new-generation SCSI interface that offers twice the speed of previous Ultra 2 SCSI, making it the highest I/O performance available.

The LaCie 18GB 15,000 rpm is the latest addition to the recently increased LaCie U160 range. By switching most of its SCSI hard drives to the new Ultra160 interface at no extra cost LaCie met the need of professionals who demanded faster hard disks. LaCie hard disk drives are now able to transfer data to and from Ultra160 SCSI controllers at 160 MB/sec. Offering a transfer rate of up to 75 MB/sec in RAID and 47 MB/sec for a single drive, as well as an impressive 3.9 millisecond seek time the LaCie 15,000 rpm hard drive is aimed at extremely demanding environments such as high end servers and audio/video applications. The new interface, coupled with the drivesU performance offers immediate benefits in throughput intensive applications where speed is the primary goal. LaCie’s Ultra 160 range is ideally suited for high traffic environments such as digital video environments, music-editing and desktop-publishing programs.


Fountain Valley, CALIF. — Kingston Technology Company, Inc. announced the release of TurboChip 566, a processor upgrade solution for 233MHz or higher Pentium II processor based systems. Designed specifically for desktop or mini-tower computers with a Slot 1 design running at 233MHz or higher, the Celeron based TurboChip 566 is an inexpensive and easy way to increase a system’s processing performance up to ninety-one percent. The TurboChip 566 is easily installed into any Slot 1 layout with 66MHz front side bus.

“With the release of the TurboChip 566, Kingston is addressing an aging 66MHz motherboard market,” said Andrew Goldberg, product manager, Peripheral Products Division, Kingston. “The evolution of processor designs have led to the development of the socket 370. With the natural migration to the socket 370, existing Slot 1 motherboards now have the inability to be upgraded. It is Kingston’s goal to prolong the life and maximize the performance of a user’s system, hence the introduction of the TurboChip 566,” added Goldberg. Available immediately, TurboChip 566, part number: (TC566/S1) carries a manufacturer’s suggested retail price of $239.00. The complete upgrade package comes with a low-profile heat sink/fan integrated cooling unit, a pass-through power cord and 2.2V to 3.3V voltage regulator. TurboChip 566 is offered through Kingston’s traditional channel of personal computer distributors and resellers. Kingston’s TurboChip566 is individually tested and backed by a lifetime warranty.


San Jose, CALIF. — Sony Electronics announced the introduction of its value-line AIT-1 (SDX-400C) series, packing a capacity of 35GB (90GB compressed) per cartridge with a data transfer rate of 4MB/sec (10MB/sec compressed). The new SDX-400C models offer scalable storage solutions with higher performance than the previous AIT-1 drives, at a lower price point. “We are pleased to offer the value-line AIT-1 solution to meet customers’ expectations for more cost-effective data protection and archival storage requirements,” said John Woelbern, senior marketing manager of tape streamer products for Sony Electronics’ Component Company. “The SDX-400C series meets the storage demands of price-sensitive customers who need a demonstrated, scalable tape technology that replaces the DDS roadmap and offers more capacity, performance and robustness than DDS. The new AIT-1 models, using Sony’s existing AME media technology, meet this requirement and provide a convenient next step up from DDS-4.”

Utilizing the smaller 3.5″ drive form-factor and more convenient cartridge size than other alternatives, Woelbern believes that AIT offers a better fit for customers migrating from DDS while providing all of the benefits of a proven, mid-range tape technology family, including substantial library automation support. The scalability and internal format compatibility of AIT technology, with a new product introduction about every two years, provides a significant investment protection benefit for customers to move within the AIT product family, as capacity and performance needs change. “The new SDX-400C drives will extend the benefits of our TLS-Series AIT Tape Library Family to cost-sensitive markets that have typically purchased DDS autoloaders. Since we shipped the first AIT library in 1996, Sony’s Advanced Tape Technology and our libraries have delivered data security for thousands of customers worldwide,” said Bob Covey, vice president of marketing at Qualstar Corporation. “We are also excited about AIT’s product roadmap, reflecting Sony’s commitment to this technology.”


Paris, FRANCE — Emulex Corp., a supplier of Fibre Channel host adapters demonstrates multi-speed Fibre Channel solutions in a storage area network (SAN) at Stockage in Paris, Sept. 12-13, 2000. In the technology demonstration, located in Booth D31, the Emulex LightPulse LP9002 Fibre Channel host adapter will operate in both 2 Gb/s (Gigabits per second) and 1 Gb/s modes in the SAN, communicating at 2 Gb/s speeds with storage devices and 1 Gb/s speeds with Brocade SilkWorm Fibre Channel fabric switches.

The speed agility feature of the Emulex LP9002 allows it to be fully backward compatible with today’s 1 Gb/s Fibre Channel standard and is also capable of providing up to 400 MB/s bandwidth on a single link when operating at 2 Gb/s full-duplex. In addition, the LP9002 is fully compatible with software drivers for the market leading LP8000 and LP7000E products. The 64-bit, 66MHz PCI LP9002 supports both 1Gb/s and 2Gb/s Fibre Channel connectivity, with auto speed sensing and auto speed matching to insure complete compatibility with existing 1Gb/s Fibre Channel SANs. The LP9002 provides a unique combination of features, including switched fabric support using F_Port and FL_Port connections, full-duplex data transfers, high data integrity features, support for all Fibre Channel topologies and support for service classes 2 and 3. The LP9002 also features sophisticated hardware that provides superior performance in storage area networks, delivering low latency and high throughput in switched, arbitrated loop and clustered environments. The LP9002 is supported by a rich suite of software, providing support for storage protocol (SCSI), network protocol (IP), as well as multi-protocol (SCSI & IP) operation, enabling advanced SAN and network applications. Emulex provides a wide range of operating system support, including Windows 2000, Windows NT, Solaris, AIX, Linux and NetWare.


San Jose, CALIF. — Promise Technology, Inc. announced the release of the first Ultra ATA RAID 5 & 3 controller product line – including a six-channel RAID card supporting up to six Ultra ATA/100 drives running at up to 100MB/sec and a four-channel RAID card for Ultra ATA/66 drives. “Our ATA RAID 5 & 3 controllers offer price differences greater than 3-to-1 when compared with even inexpensive Ultra2 LVD SCSI RAID configurations,” said Sam Sirisena, Promise vice president of sales and marketing. “We have become the first and only manufacturer to provide a complete product line for network professionals seeking reliable, economically-sensible ATA RAID 0, 1, 0/1, 3 or 5 storage solutions with complete RAID features such as hot-swap and hot spare.”

“As the pioneer of the ATA RAID concept and the acknowledged industry leader of this product category, we are dedicated to assuring that consumers continue to have a complete and comprehensive selection of ATA RAID controllers and subsystems that offer SCSI RAID functions at one third of SCSI pricing.” The company’s SuperTrak100 (for Ultra ATA/100 drives) and SuperTrak66 (for Ultra ATA/66 drives) RAID cards with onboard i960RD processors and cache memory are designed for small to midrange-level networks with easy initial setup and across-network array management tools dedicated to ATA RAID. Each SuperTrak product offers complete fault tolerance options with choice of RAID 1, 0+1, 3, or 5 levels, “hot” swap, and “hot” spare support. The cards work seamlessly with Promise Technology’s optional SuperSwap removable drive housing and chassis for “hot” swap operation and enclosure management. LEDs indicate drive and array status – including disk access, disk failure, and disk rebuild. In addition, the cards’ advanced array management GUI keeps track of each SuperSwap enclosure by monitoring temperature, fan status, and voltage regulation.


Bridgewater, N.J. — StorageApps Inc., a leading provider of all-encompassing solutions, applications and appliances for storage-based network services, announced that it had extended its Proactive Network Storage Solutions offerings to provide customers with a smarter way to implement Network Attached Storage: the SANLink Plus appliance. Network Attached Storage (NAS) is one of the approaches companies are using to deal with the incredible explosion of e-business data. NAS attaches directly to the local area network (LAN) using traditional protocols such as Ethernet and TCP/IP and serves files to any host or client also connected to the LAN. According to Steve Duplessie, an analyst at Enterprise Storage Group, NAS is the logical approach for storing files that must be shared by standard office productivity applications that execute on a desktop, in part because of its simplicity. “You plug it in, give it an IP address, and all your heterogeneous hosts know how to deal with it,” Duplessie said.

“Our customers have been telling us is that they like the way NAS can be added to their networks without the need to implement and manage a new protocol, such as Fibre Channel,” said Dan Petrozzo, president and chief operating officer of StorageApps. “But they’re concerned about the scalability, availability, functionality and management of current NAS solutions. “A number of these clients asked us to help them find a new approach to NAS, a smarter approach that would maintain the simplicity of NAS while providing the flexibility, availability and growth options required in a rapidly evolving e-business environment. We’ve responded by adding NAS to our Proactive Network Storage Solutions family of products.” For more information on StorageApps and its Proactive Network Storage Solutions portfolio of products, visit the company at .


San Diego, CALIF. — Maxoptix Europe Ltd, ( ), a leader in optical data- storage technology, introduces two DVD-RAM libraries, marking the company’s entrance into the market. The Maxoptix DVD520 and DVD540 are the only DVD-RAM jukeboxes in the industry designed specifically to meet the random-access requirements of DVD-RAM – and the only jukeboxes that can use the standard DVD cartridges approved and recommended by the DVD Forum. (All other DVD-RAM libraries are adaptations of products originally designed for CDs – and can handle only single-sided bare media.) The DVD520 and DVD540 jukeboxes provide a cost-per-gigabyte comparable to tape, with much faster access – and a ruggedness and archivability comparable to magneto-optical (MO). “The DVD520 and DVD540 provide excellent media durability, because they use standard DVD-RAM cartridges,” said Brian Blanchard, Maxoptix Managing Director. “Since the media is never touched by human hands, it cannot be contaminated by dust or from handling. Over time, repeated handling of unprotected rewritable media can cause significant contamination.

Fingerprints are especially problematic, because they degrade the signal level reaching the disc’s recording layer – cartridged media completely eliminates this problem.” Media cost is also a concern with libraries. Maxoptix jukeboxes use standard double-sided cartridges, which cost about half as much as the media used by competitive units, at GBP27 per media cartridge. This can result in cost savings of several hundred pounds per jukebox. The DVD520 and DVD540 use 5.2GB double-sided (2.6GB per side) DVD-RAM cartridges. This gives the 20-cartridge DVD520 a capacity of 104GB and the 40- cartridge DVD540 a capacity of 208GB. The jukeboxes feature a flipping mechanism to take advantage of the storage capacity on both sides of the discs, and are available with one or two drives.


Atlanta, GA — Continuing to raise the standards and level of expectations in the telecom industry, telecom technologies, inc., The Softswitch Source announced from Fall VON 2000 the creation of its INIP Powered Virtual Lab, an online resource for real-time status of the INIP Powered Interoperability Lab. The Virtual Lab can be visited at . The INIP Powered Interoperability Lab is a one-of-a-kind facility designed to stimulate interoperability between vendors’ products and softswitch technology. Within the lab, partners undergo testing for interoperability with the INtelligentIP(TM) softswitch in order to provide carriers with products that can optimize on technology within next generation networks. The lab is more than 1450 square feet of environmentally controlled facilities located in the heart of the Richardson, Texas Telecom Corridor.

Targeted toward carriers and vendors, the Virtual Lab will enable visitors to gather information from the INIP Powered Interoperability Lab, as well as actually see into the lab from anywhere in the world. Carriers, vendors and members of the media will be able to learn which partners are currently testing, which products are being demonstrated in the lab, which partners and products are INIP Powered, which partners are currently in customer trial testing, as well as which partners have participated in integrated solutions and the network architecture design for those integrated solutions. “We give carriers insight into what we and other industry leaders like Cisco Systems, Lucent and Tellabs, as well as such emerging vendors as LongBoard and BroadSoft are doing in the lab. We are not promoting hype. This will be a site that carriers, vendors, and members of the media will want to bookmark to keep up with the progress of interoperability in the industry,” said Anousheh Ansari, chairman and CEO of telecom technologies.


Austin, TEXAS — HyPerformix Inc., a leader in solving performance problems, announced the availability of a Windows NT version of its popular discrete-event simulation modeling tool SES/workbench and the availability of the latest version of its enterprise simulation modeling tool, Strategizer. “These product releases reinforce our commitment to predictive simulation modeling technology,” stated Noel Barnard, President and CEO of HyPerformix. “When combined with our performance services, these products provide the solution eBusiness customers require in order to minimize the risk of poor performance.” SES/workbench is a visual simulation environment used to model complete systems for performance analysis and functional verification. SES/workbench is well suited for specifying and evaluating complex systems involving a high-degree of concurrent processing, such as computer hardware and software systems, electronic components, computer modules, communication systems, and manufacturing systems.

“As a long-time user of SES/workbench on a Unix platform we are excited about having the product run on NT,” stated Robert Graham, senior engineer at Lockheed Martin. “This product is a welcome addition to our NT environment and plays an important role in assuring the performance of our products.” Hundreds of global 2000 companies use SES/workbench to validate the performance of components and systems before they ever enter production. The performance assurance provided by SES/workbench saves millions of dollars in design and testing costs as well as preventing loss of critical time needed to correct problems.


St. Louis, MO — Unigraphics Solutions Inc. announces the Parasolid Innovator program, making Parasolid, UGS’ industry-leading component geometric modeler, available to small innovative companies under business terms focused on their specific needs. With this new program, Parasolid, the engine used to power the leaders of the current generation of CAD/CAM/CAE products, will now be the component modeler of choice for developers of the next generation systems. More than two years ago Unigraphics Solutions pioneered the pay-as-you-go pricing model in the component geometric modeler market by eliminating all up front license charges. As a result of this revolutionary approach to pricing, many excellent business relationships were formed. Now with the Innovator program, Unigraphics Solutions takes the risk and reward sharing philosophy to the next level, by offering smaller companies the opportunity to develop against Parasolid with no financial commitments for up to one full year.

History shows that Parasolid’s easy to use technology enables adopters to get profitable products to market quickly and with relatively little effort, enabling small companies with good ideas and talented people to make an impact in the MCAD industry. By encouraging these companies, Unigraphics Solutions expects to invigorate the development of innovative MCAD products and further expand the community of 600,000 active end-users of Parasolid-powered products. For more details visit . “After they have completed the development of their products many Parasolid customers comment that the process was much easier than they had anticipated,” says Ken Sears, leader of the Parasolid Line of Business. “The objective of the Innovator program is to build on this observation by providing an incubation environment for talented people in small companies aiming to develop new products for the mechanical CAD/CAM/CAE segments. We expect this will be a very positive experience for fledgling companies and we are confident that our investment in them will provide a worthwhile return for both parties.”


San Jose, CALIF. — Cadence Design Systems, Inc., a leading supplier of electronic design products and services, announced that it would make its TestBuilder C++ testbench class library available through open source licensing – the first such tool to be available on this basis. The company will provide free and open access to its TestBuilder library – as well as on-going source code management – as the community of developers grows. This move will allow designers, IP developers, and EDA vendors to develop interoperable testbenches for chip or system design verification. In addition, Cadence announced the initial participants in a Developers Forum, which includes Sun Microsystems, CynApps, and ASIC Alliance. Developers Forum participants receive a high level of support and direct access to the TestBuilder development team.

TestBuilder provides testbench authoring capabilities to create reusable, random, constraint-driven, self-checking tests. Based on standard C++ programming language, it is easy to learn and provides the power and flexibility of object-oriented techniques. It provides support for a transaction-based verification methodology in which tests can be created at the architectural level, and allows the development of sophisticated testbenches without the need for custom interfaces to simulators. TestBuilder’s integration with Cadence’s Verification Cockpit provides users with a complete environment that includes simulation, transaction analysis, code coverage, and HDL analysis. Cadence will also support a variety of vendor’s simulators, emulators and accelerators, in addition to its own NC-Simulator. “The goals of the open source model are to maximize the reuse of verification-related intellectual property (IP), and to create a fully functional standard through interaction and cooperation between a community of users, ” said Rahul Razdan, vice president for Front-End Products at Cadence. “This approach has proven itself to be a viable, valuable way for specific communities to grow and benefit from interaction. We are happy to take a leading role in developing a community of users in the area of testbench authoring.” For more information, visit .


Burlington, MASS. — UPSPRING Software, Inc., a technology leader in software infrastructure analysis solutions, announced the immediate availability of CodeRover Browser, a stand-alone desktop source code browser designed to enhance and accelerate an individual developer’s understanding and navigation of source code applications. CodeRover Browser is one of eight new CodeRover applications from UPSPRING that enhance the productivity of individual software developers. CodeRover Browser is available for download at a special introductory price of $195 by visiting UPSPRING’s Web site at . To enable rapid navigation, the CodeRover Browser captures relationships between source code entities in an accurate and complete personal database for any combination of C, C++, Java, and SQL languages. A powerful query engine, a series of over 50 standard queries, plus graphical and textual views, help software developers to rapidly comprehend and enhance software applications more efficiently than ever before. Integration with common IDE’s such as vi, emacs, and Microsoft Visual Studio allow changes and additions to be made swiftly and accurately. Ultimately, the CodeRover Browser provides a newly assigned developer with the ability to quickly understand a project and reduces the need to consult with others, thus increasing personal productivity.

At the heart of every CodeRover application is a comprehensive personal database built by automatically scanning the source code and capturing details describing all entities in a developer’s project, including macros and difficult implicit compiler-generated programming constructs, plus their inter-relationships. It is the depth and accuracy of the database, as well as the intelligence of relationships, which help CodeRover applications to provide performance superior to other products on the market today. “We tested the CodeRover Browser on a real project where our task was to port a large body of C++ code to Java,” said Boris Furman, Senior Software Engineer at Vignette Corporation. “The code contained complicated inheritance relationships, used macros to hide many details and was generally hard to understand. Using the CodeRover Browser, we were able to gain necessary insight into the code much more rapidly and thoroughly than we would have with other tools we investigated. We got instantaneous answers to queries that would have required hours with other tools.” Utilizing the multi-language CodeRover database as a starting point, a developer uses the CodeRover Browser to get a quick overview of a software project. CodeRover Browser provides a hierarchical view by listing all the entities sorted by attribute, including; name, size, cyclomatic complexity, or other similar characteristics, contained within a project. To learn more about specific entities, the CodeRover Browser provides graphical views showing the relationship between entities. A Function Call Tree diagram provides a cascading view of what a function calls. Additionally, the diagram lists all the functions that call the function of interest. Class Inheritance diagrams produce a “family tree” for a selected class, displaying all of its subclasses (children) and superclasses (parents).


Paris, FRANCE — Plasmon, the leading provider of managed removable storage, has expanded its tape storage offering with the launch of a new 1/2 inch linear tape automation range. Utilising LTO-Ultrium and SuperDLT technology, the new tape range will not only offer users a wider choice in storage technology from Plasmon, but considerable cost savings compared to similar products in the market. The new range will provide customers with unprecedented value providing low cost, high quality storage functionality. The new tape products compliment Plasmon’s MP Series of tape automation and position Plasmon as not only a provider of complete storage solutions also including RAID, CD, DVD, 12″ TrueWORM Optical and Magneto-Optical, but one whose comprehensive tape range offers the latest in tape storage technology.

Plasmon’s 1/2 inch linear tape is suitable for most tape storage applications including back up and recovery, the creation of archive files and data warehousing. The tape can accommodate the storage of large sequential data sets and hierarchical storage. Because tape is a removable technology, it means that off-site storage requirements can be satisfied in-line with an organisation’s disaster recovery policy. This makes Plasmon’s LTO-Ultrium and SuperDLT tape products an ideal solution for back-up storage on distributed systems. The new range will be based on IBM’s StorageSmart Ultrium tape drives and will also support both HP and Seagate solutions to offer extra flexibility to customers. Tape cartridges will be available in 10, 30, 50 and 100 Gigabyte (native) capacity and an initial data recovery rate of approximately 15 megabytes per second. David Brocklebank, vice president of sales and marketing EMEA at Plasmon comments: “We recognise that with the emergence of the new competing LTO- Ultrium and SuperDLT technologies, customers are faced with a storage choice. That’s why our new 1/2 inch linear tape range utilises LTO-Ultrium and SuperDLT technology – providing a future-proof tape storage solution. End-users can be confident that by choosing Plasmon’s 1/2 inch linear tape, their storage and business needs will be supported both now and in the future.”


Sunnyvale, CALIF. — Network Appliance, Inc., a leading provider of network-attached data access and content management solutions, unveiled its new system platform specifically designed to dramatically increase the scalability, reliability, and performance of global storage infrastructures. The F840c and F840, combined with the next generation of Network Appliance’s robust operating system Data ONTAP 6.0, are the first products enabling data access and content delivery solutions to scale from tens to hundreds or even thousands of terabytes of capacity based on interoperable appliance building blocks. They are also the first products unveiled in Network Appliance’s “end-to-end” data management and content delivery strategy, which was announced in a separate press release on September 5, 2000.

Network Appliance developed this new system platform to address customer requirements for far larger and more robust storage network solutions. These products offer customers significantly greater capacity than previous NetApp solutions, and provide levels of data availability unmatched by any other storage network vendor. The platform is especially appealing for large global enterprises, content delivery networks (CDNs), Internet service providers (ISPs), application service providers (ASPs), and storage service providers (SSPs). In effect, Network Appliance has dramatically changed how storage networks are deployed, enabling those customers with massive capacity environments to simplify, share, and scale their critical storage networking infrastructures. As an added benefit, customers can also substantially increase the performance of their mission-critical business applications such as database, collaborative, and home directory applications. In industry standard CIFS and NFS benchmark tests (NetBench 6.0 and SPEC SFS97), the F840c and F840 are the industry’s highest performing dual- and single-processor configurations, ranking 2-3 times better performance in throughput and response time than competitors’ offerings. Furthermore, these RAID-protected, single file system, results exhibit leading price/performance and lowest total cost of ownership in contrast to more complex systems.


Research Triangle Park, N.C. — Red Hat, Inc., a leading provider of open source Internet infrastructure, announced advanced technical training courses focusing on embedded systems engineering, IA-64 engineering, and Apache and secure Web server administration. Recent acquisitions and the scope of Red Hat’s capabilities have expanded to include a full spectrum of solutions for embedded devices such as Internet appliances and routers to secure e-commerce applications.

“Red Hat is in the business of enabling users to leverage powerful open source technologies and solutions throughout the ecosystem of devices and applications that connect to and power the Internet,” said Paul McNamara, vice president of products and platforms, Red Hat. “These new courses are further steps in our total commitment to support customers, partners, and developers as they work with open source technology on everything from Internet appliances, to server and Internet infrastructures, and new platforms such as IA-64,” said Michael Tiemann, chief technology officer, Red Hat. “Advanced training at Red Hat is designed to meet the needs of individuals and engineering teams as they build new devices and applications, and manage secure network services on the latest hardware.” The Red Hat Embedded Systems Engineering Course teaches embedded developers how to use the Red Hat Embedded DevKit and Red Hat GNUPro to develop, build, test, and debug embedded Linux applications. Labs are emphasized throughout, and each participant spends the final day actually developing an embedded application for a small system on a reference board. The Red Hat IA-64 Engineering Seminar features hands-on installation and configuration of Linux on IA-64 machines, and discussion of key differences in systems administration and development for Linux on the new Intel 64-bit architecture. Red Hat is the first company to offer regularly scheduled training on Linux running on the new Intel IA-64 architecture.


Detroit, MICH. — PlanetCAD, a division of Spatial Inc., announced the release of version 4 of DesignQA from PrescientQA’s premier family of design-quality software. DesignQA is a leading engineering quality assurance software for manufacturers that detects, assesses, corrects and prevents product development problems caused by inaccurate, incomplete or inconsistent design modeling practices. This quality assurance tool helps ensure 3D models: are reusable by other CAD designers; can incorporate data requirements of downstream applications; are usable by analysis, manufacturing and rapid prototyping systems; and are prepared for PDM processes. DesignQA is used by many Fortune 500 companies in aerospace, automotive, electronics and other discrete manufacturing industries.

The PrescientQA premier software family includes DriveQA, GeometryQA, CertifyQA software and the AuditQA quality audit. PrescientQA is the only software suite available that provides multi-CAD system detection, correction, improvement, management, and control of all aspects of quality in the product definition process. Wade Malwitz, a member of PlanetCAD’s customer advisory board for PrescientQA, has been beta testing DesignQA version 4 at a major transportation manufacturing company. Wade remarked, “The integration and interaction with CATIA(R), plus improved performance and functionality, make DesignQA version 4 a strategic tool for improving our product quality and engineering process.” DesignQA version 4 will work and interact with CATIA, Pro/ENGINEER, and Unigraphics. For more information about PrescientQA products, visit .


Cupertino, CALIF. — Intuitive Systems, Inc. today announced shipment of Optimizeit 4.0, the latest version of its comprehensive Java technology-based language profiling tool that allows developers to test and improve the performance of any Java programs including servlets, Enterprise JavaBeans, Java Server Pages and complex server-based Java applications. New features include universal Java virtual machine support, offline profiling for testing of programs in production environments, and integration with even more application servers.

Optimizeit enables Java technology developers to understand and solve performance issues such as memory leaks, excessive object allocation, inefficient algorithms and inefficient processor use in their Java programs. Optimizeit combines scalability and ease-of-use in a complete profiling solution to provide the most comprehensive and accurate information on how Java programs use CPU and memory. Because it instantly integrates with most popular application servers, Optimizeit enables developers to profile their Web applications and track potential performance problems quickly and easily. “We’re pleased to see high quality tools such as Optimizeit use the Java technology from IBM for Windows and Linux,” said Rod Smith, vice president of Java Software, IBM. “Performance software like Optimizeit coupled with IBM’s Java runtimes provide developers with powerful tools for deploying e-business applications.”


Bothell, WASH. — Vixel Corporation announced that it has been awarded a United States Patent relating to its private loop switching technology. U.S. Patent No. 6,118,776, entitled “Methods and Apparatus for Fibre Channel Interconnection of Private Loop Devices,” defines a number of ways to interconnect private loop devices with a Fibre Channel switch. Fibre Channel standards define multiple topologies including point to point, arbitrated loop and full fabric, but virtually all of the initial and current deployments of the technology use point to point or the loop topology. Since arbitrated loop is less complex to deploy and manage than a full fabric topology, many computer systems vendors only support arbitrated loop implementations. Support for full fabric is now just beginning to be deployed. “The technology was conceived by the requirement of end users to maintain the simplicity of the loop and at the same time enjoy the benefits of switch functions,” commented Stuart Berman, Vixel’s Chief Technology Officer and the patent’s author. “The patent defines ways to interconnect devices through a switch without the complexities of full fabric services support. Because of the simplicity of interconnection that this technology provides, private loop switching is utilized in a majority of both new and existing Storage Area Network (SAN) installations.”

“We are delighted to have received recognition from the United States Patent Office on key technology we initially pioneered,” commented Jim McCluney, Vixel’s President and CEO. “This patent, the first of several additional pending patents, highlights Vixel’s industry leading Fibre Channel switch technology. Private loop switching technology has been implemented in Vixel’s complete line of fabric products and was key in establishing Vixel as the leading switch provider to the digital video marketplace.” Using this technology, Vixel has developed a switch product family that provides users with a cost-effective alternative that provides the ease of use of arbitrated loops and the high performance of switching environments. As SANs expand, users can easily migrate the private loop switching environments to full fabric services by a simple firmware upgrade.


Bristol, ENGLAND — Intel Corporation announced that it has released the first Linux version of its Open Source Computer Vision Library. The Open Source Computer Vision Library provides a wide range of functions, including gesture recognition, object-tracking, face recognition and camera calibration, that will help researchers develop ways to use “computer vision” as a method for people to interact with computers. The announcement took place at the opening of the industry day events at the eleventh British Machine Vision Conference. Earlier this year, Intel released a computer vision library and its source code for Microsoft Windows. Since the Open Source Computer Vision Library was first announced at the IEEE Computer Society Conference on Computer Vision and Pattern Recognition in June of this year, over 500 researchers worldwide have joined Web-based computer vision user groups.

Intel will facilitate the growth and maintenance of this library by accepting and maintaining submissions of source code content reviewed by a panel of experts from the computer vision community. Reviewers include representatives from major vision laboratories. This open source software release includes C source code for all of the library’s functions and a royalty-free redistribution license. The library may be downloaded from .


Albuquerque, N.M. — Innovative software systems designed to enhance air traffic management and a wide range of aircraft training functions will be demonstrated by MUSE Technologies at the InterAirport Atlanta trade show, being held September 19-21, 2000 at the Georgia International Convention Center in Atlanta, Georgia. Recent projects by MUSE Technologies to be demonstrated at the Atlanta show include a prototype air traffic management system developed by MUSE for the John A. Volpe National Transportation Systems Center and an avionics training program that recreates the cockpit of a RAF Tornado F3 jet fighter using highly interactive 3D graphics and virtual reality. The latter project was developed with Alenia Marconi Systems for the Royal Air Force and has produced significant increases in student comprehension and training efficiency.

MUSE will demonstrate its software on the Panoram PV-290 multi-screen desktop display, which fuses three high-resolution side-by-side monitors into a single, small footprint device. The demonstration of MUSE software and Panoram display technology will provide air industries with a compelling example of how 3D graphics are poised to bring new insight to all types of industry-specific data. MUSE Technologies, Inc. is a developer of advanced perceptual computing and visualization software and solutions. From headquarters in Albuquerque, NM and offices in Houston, Washington D.C., London, Manchester (UK) and Paris, the Company and its subsidiaries, Virtual Presence and MUSE Federal Systems Group, provide a broad range of software, hardware, integration and development solutions to corporate and government customers in an array of industries. For more information on MUSE Technologies, visit the company’s Web site at .


Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

What’s New in Computing vs. COVID-19: Fugaku, Congress, De Novo Design & More

July 2, 2020

Supercomputing, big data and artificial intelligence are crucial tools in the fight against the coronavirus pandemic. Around the world, researchers, corporations and governments are urgently devoting their computing reso Read more…

By Oliver Peckham

OpenPOWER Reboot – New Director, New Silicon Partners, Leveraging Linux Foundation Connections

July 2, 2020

Earlier this week the OpenPOWER Foundation announced the contribution of IBM’s A21 Power processor core design to the open source community. Roughly this time last year, IBM announced open sourcing its Power instructio Read more…

By John Russell

HPC Career Notes: July 2020 Edition

July 1, 2020

In this monthly feature, we'll keep you up-to-date on the latest career developments for individuals in the high-performance computing community. Whether it's a promotion, new company hire, or even an accolade, we've got Read more…

By Mariana Iriarte

Supercomputers Enable Radical, Promising New COVID-19 Drug Development Approach

July 1, 2020

Around the world, innumerable supercomputers are sifting through billions of molecules in a desperate search for a viable therapeutic to treat COVID-19. Those molecules are pulled from enormous databases of known compoun Read more…

By Oliver Peckham

HPC-Powered Simulations Reveal a Looming Climatic Threat to Vital Monsoon Seasons

June 30, 2020

As June draws to a close, eyes are turning to the latter half of the year – and with it, the monsoon and hurricane seasons that can prove vital or devastating for many of the world’s coastal communities. Now, climate Read more…

By Oliver Peckham

AWS Solution Channel

Maxar Builds HPC on AWS to Deliver Forecasts 58% Faster Than Weather Supercomputer

When weather threatens drilling rigs, refineries, and other energy facilities, oil and gas companies want to move fast to protect personnel and equipment. And for firms that trade commodity shares in oil, precious metals, crops, and livestock, the weather can significantly impact their buy-sell decisions. Read more…

Intel® HPC + AI Pavilion

Supercomputing the Pandemic: Scientific Community Tackles COVID-19 from Multiple Perspectives

Since their inception, supercomputers have taken on the biggest, most complex, and most data-intensive computing challenges—from confirming Einstein’s theories about gravitational waves to predicting the impacts of climate change. Read more…

Hyperion Forecast – Headwinds in 2020 Won’t Stifle Cloud HPC Adoption or Arm’s Rise

June 30, 2020

The semiannual taking of HPC’s pulse by Hyperion Research – late fall at SC and early summer at ISC – is a much-watched indicator of things come. This year is no different though the conversion of ISC to a digital Read more…

By John Russell

OpenPOWER Reboot – New Director, New Silicon Partners, Leveraging Linux Foundation Connections

July 2, 2020

Earlier this week the OpenPOWER Foundation announced the contribution of IBM’s A21 Power processor core design to the open source community. Roughly this time Read more…

By John Russell

Hyperion Forecast – Headwinds in 2020 Won’t Stifle Cloud HPC Adoption or Arm’s Rise

June 30, 2020

The semiannual taking of HPC’s pulse by Hyperion Research – late fall at SC and early summer at ISC – is a much-watched indicator of things come. This yea Read more…

By John Russell

Racism and HPC: a Special Podcast

June 29, 2020

Promoting greater diversity in HPC is a much-discussed goal and ostensibly a long-sought goal in HPC. Yet it seems clear HPC is far from achieving this goal. Re Read more…

Top500 Trends: Movement on Top, but Record Low Turnover

June 25, 2020

The 55th installment of the Top500 list saw strong activity in the leadership segment with four new systems in the top ten and a crowning achievement from the f Read more…

By Tiffany Trader

ISC 2020 Keynote: Hope for the Future, Praise for Fugaku and HPC’s Pandemic Response

June 24, 2020

In stark contrast to past years Thomas Sterling’s ISC20 keynote today struck a more somber note with the COVID-19 pandemic as the central character in Sterling’s annual review of worldwide trends in HPC. Better known for his engaging manner and occasional willingness to poke prickly egos, Sterling instead strode through the numbing statistics associated... Read more…

By John Russell

ISC 2020’s Student Cluster Competition Winners Announced

June 24, 2020

Normally, the Student Cluster Competition involves teams of students building real computing clusters on the show floors of major supercomputer conferences and Read more…

By Oliver Peckham

Hoefler’s Whirlwind ISC20 Virtual Tour of ML Trends in 9 Slides

June 23, 2020

The ISC20 experience this year via livestreaming and pre-recordings is interesting and perhaps a bit odd. That said presenters’ efforts to condense their comments makes for economic use of your time. Torsten Hoefler’s whirlwind 12-minute tour of ML is a great example. Hoefler, leader of the planned ISC20 Machine Learning... Read more…

By John Russell

At ISC, the Fight Against COVID-19 Took the Stage – and Yes, Fugaku Was There

June 23, 2020

With over nine million infected and nearly half a million dead, the COVID-19 pandemic has seized the world’s attention for several months. It has also dominat Read more…

By Oliver Peckham

Supercomputer Modeling Tests How COVID-19 Spreads in Grocery Stores

April 8, 2020

In the COVID-19 era, many people are treating simple activities like getting gas or groceries with caution as they try to heed social distancing mandates and protect their own health. Still, significant uncertainty surrounds the relative risk of different activities, and conflicting information is prevalent. A team of Finnish researchers set out to address some of these uncertainties by... Read more…

By Oliver Peckham

[email protected] Turns Its Massive Crowdsourced Computer Network Against COVID-19

March 16, 2020

For gamers, fighting against a global crisis is usually pure fantasy – but now, it’s looking more like a reality. As supercomputers around the world spin up Read more…

By Oliver Peckham

[email protected] Rallies a Legion of Computers Against the Coronavirus

March 24, 2020

Last week, we highlighted [email protected], a massive, crowdsourced computer network that has turned its resources against the coronavirus pandemic sweeping the globe – but [email protected] isn’t the only game in town. The internet is buzzing with crowdsourced computing... Read more…

By Oliver Peckham

Global Supercomputing Is Mobilizing Against COVID-19

March 12, 2020

Tech has been taking some heavy losses from the coronavirus pandemic. Global supply chains have been disrupted, virtually every major tech conference taking place over the next few months has been canceled... Read more…

By Oliver Peckham

Supercomputer Simulations Reveal the Fate of the Neanderthals

May 25, 2020

For hundreds of thousands of years, neanderthals roamed the planet, eventually (almost 50,000 years ago) giving way to homo sapiens, which quickly became the do Read more…

By Oliver Peckham

DoE Expands on Role of COVID-19 Supercomputing Consortium

March 25, 2020

After announcing the launch of the COVID-19 High Performance Computing Consortium on Sunday, the Department of Energy yesterday provided more details on its sco Read more…

By John Russell

Steve Scott Lays Out HPE-Cray Blended Product Roadmap

March 11, 2020

Last week, the day before the El Capitan processor disclosures were made at HPE's new headquarters in San Jose, Steve Scott (CTO for HPC & AI at HPE, and former Cray CTO) was on-hand at the Rice Oil & Gas HPC conference in Houston. He was there to discuss the HPE-Cray transition and blended roadmap, as well as his favorite topic, Cray's eighth-gen networking technology, Slingshot. Read more…

By Tiffany Trader

Honeywell’s Big Bet on Trapped Ion Quantum Computing

April 7, 2020

Honeywell doesn’t spring to mind when thinking of quantum computing pioneers, but a decade ago the high-tech conglomerate better known for its control systems waded deliberately into the then calmer quantum computing (QC) waters. Fast forward to March when Honeywell announced plans to introduce an ion trap-based quantum computer whose ‘performance’ would... Read more…

By John Russell

Leading Solution Providers


Neocortex Will Be First-of-Its-Kind 800,000-Core AI Supercomputer

June 9, 2020

Pittsburgh Supercomputing Center (PSC - a joint research organization of Carnegie Mellon University and the University of Pittsburgh) has won a $5 million award Read more…

By Tiffany Trader

‘Billion Molecules Against COVID-19’ Challenge to Launch with Massive Supercomputing Support

April 22, 2020

Around the world, supercomputing centers have spun up and opened their doors for COVID-19 research in what may be the most unified supercomputing effort in hist Read more…

By Oliver Peckham

Australian Researchers Break All-Time Internet Speed Record

May 26, 2020

If you’ve been stuck at home for the last few months, you’ve probably become more attuned to the quality (or lack thereof) of your internet connection. Even Read more…

By Oliver Peckham

15 Slides on Programming Aurora and Exascale Systems

May 7, 2020

Sometime in 2021, Aurora, the first planned U.S. exascale system, is scheduled to be fired up at Argonne National Laboratory. Cray (now HPE) and Intel are the k Read more…

By John Russell

Nvidia’s Ampere A100 GPU: Up to 2.5X the HPC, 20X the AI

May 14, 2020

Nvidia's first Ampere-based graphics card, the A100 GPU, packs a whopping 54 billion transistors on 826mm2 of silicon, making it the world's largest seven-nanom Read more…

By Tiffany Trader

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

By Doug Black

Summit Supercomputer is Already Making its Mark on Science

September 20, 2018

Summit, now the fastest supercomputer in the world, is quickly making its mark in science – five of the six finalists just announced for the prestigious 2018 Read more…

By John Russell

TACC Supercomputers Run Simulations Illuminating COVID-19, DNA Replication

March 19, 2020

As supercomputers around the world spin up to combat the coronavirus, the Texas Advanced Computing Center (TACC) is announcing results that may help to illumina Read more…

By Staff report

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This