HPCwire: Etnus Relaunches

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

Leading HPC
Solution Providers

HPCwire >> Features

April 13, 2007

Etnus Relaunches

by Michael Feldman, HPCwire Editor

Etnus LLC, makers of the TotalView debugger, have apparently decided to worship at the altar of name recognition. Last week, they relaunched the company under a new name, TotalView Technologies, using the moniker of their flagship debugger offering. The new company also announced TotalView Multi-Core Debugging Framework, a new workbench model that revolves around their core debugging technology. The framework is designed to integrate a set of debugger and analysis tools that will help developers deal with parallel programming applications targeted for multi-core architectures.

"For years our customers have recognized us as the TotalView company," explains Rich Collier, CEO of TotalView Technologies. "So we just went with the flow. It also allows us to capitalize on the success of our flagship product and it ties in with our other new product that we're going to be coming out with underneath the TotalView umbrella."

The TotalView debugger was originally developed at BBN (Bolt Beranek and Newman) for their Butterfly computer, a massively parallel system built in the 1980s. In its day, the Butterfly was a state-of-the-art machine, and was probably as challenging to program as IBM Cell-based systems are today. The Butterfly is no more, but the BBN ToolWorks group took the TotalView technology and spun off a new company, Etnus, in 1999. Over the next eight years, TotalView went on to become the most widely known commercial parallel debugger for high performance machines.

In the early part of the decade, the TotalView source code debugger started out in government labs and supercomputer centers, where most of the HPC action was. But starting in 2001, Etnus began to tap into the growing number of commercial customers as they adopted parallel programming technologies.

"Today TotalView is used not just for high-end applications, but also for distributed, cluster, or any heavily multi-threaded codes," says Collier.

It's being used to debug everything from nuclear weapons simulation programs to financial analysis codes to animation rendering workloads. With regards to animation, Collier says that their technology has become the de facto standard for all animation and digital content studios. Recently, Blue Sky Studios used TotalView during the production of the animated film "Ice Age." Disney and Pixar are also using the debugger for their computer-generated animation projects.

TotalView is well-known for supporting a wide variety of hardware and operating systems found on HPC platforms. The breadth of coverage includes Blue Gene/L (PowerPC) on SuSE Linux, Itanium 2 on HP-UX, SPARC on Solaris, Intel x86 on Mac OS X, to name a few. Collier says they're also considering support for Microsoft's Windows Compute Cluster Server (CCS) platform. According to him, they've received a few requests from some existing customers and have also talked to partners about it, but haven't made a commitment yet. No doubt they'll hop on board if and when CCS gains enough traction in the HPC market to make it worthwhile.

The new framework announced last week encompasses a set of five debugging and analysis components that can be incorporated into a workbench structure. The components include:

TotalView Source Debugger -- the core technology for source code debugging (just updated)
MemoryScape Memory Debugger -- for finding memory leaks or data corruption in dynamic (heap) memory
Performance Analysis Tool Suite -- for performance tuning
Data-Centric Debugging Tool Suite -- for detecting data errors in a multi-threaded environment
Active Web Tool Debugging Suite -- allows developers to debug in distributed Java and AJAX environments

The three tool suites won't be available until the second half of 2007. The other two components are available now and amount to a repackaging of existing offerings. Up until now, TotalView included source code and memory debugging together. With the memory debugging broken out in the MemoryScape product, this can be purchased and licensed separately.

Although TotalView has few parallel debugger competitors (Allinea's Distributed Debugging Tool being one), a lot of users are still using GDB, or variations of it, to debug their HPC applications. In general, GDB does only serial debugging, so does not offer the sophistication of an MPI- or OpenMP-aware parallel debugger. Some users prefer to go without debuggers altogether, inserting low-level printf calls into their code to output data or other application status information during test runs. Less sophisticated methods like printf can work for smaller codes, but once the application scales up to hundreds or thousands of processors, a more high level view of the software is usually required.

"What we're doing with this framework is offering a comprehensive, integrated set of software development tools that are designed to work together to provide multi-core debugging. Ultimately this will improve the programmer's productivity and keep the programmer sane."

Selling productivity and sanity is a good strategy when you're talking about parallel debugging. Memory leaks, data corruption and logic errors all become much harder to find and fix when you start using hundreds or thousands of threads. TotalView Technologies' plans to follow parallel architectures into the marketplace and provide the debugging and analysis tools that are bound to be needed by a new generation of applications.

Article Tools

Share & Save Options

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCWire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCWire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Top Headlines

Heard at Hot Interconnects: Multicore SoCs May Require Optical Interconnect

Aug 27 | EDN | The lead-off session at the Hot Interconnects conference at Stanford examined the growing challenge of chip interconnect technology as today's multicore processors evolve into chips with tens or hundreds of cores. Read more...

Data Centers: Focusing on Sustainability

Aug 27 | Dr. Dobb's Portal | Dallas Thornton, director of Cyberinfrastructure Services for the San Diego Supercomputer Center, offers his thoughts on how energy concerns and other datacenter issues are affecting the industry. Read more...

China Aims for Petaflop Computer in 2010

Aug 27 | InfoWorld | China still lags behind its international rivals in chip development but is stepping up investment in its homegrown Godson microprocessor. Read more...

Really Big Things

Aug 27 | Linux Magazine | How does one manage really big clusters? Perhaps nature can give us a clue. Read more...

Fujitsu Readies Eight-Core Sparc64 Chip

Aug 27 | Computerworld | Fujitsu is developing an eight-core version of its Sparc64 processor, which should give a performance boost to the Sparc Enterprise Servers that Fujitsu jointly develops with Sun Microsystems. Read more...

Read more headlines...

Featured Whitepapers

Foundry High Performance Computing Application Note

Aug 01 | | The intent of this document is to provide an overview of Foundry Networks products that support high end HPCC configurations.

Improving Performance and Manageability for Seismic Processing and Imaging Applications with Parallel Storage

Jun 05 | | As pressure increases on the upstream seismic processing community to deliver ever-higher levels of productivity and efficiency, a new generation of storage solutions will be required that allow the maximum utilisation of high-performance computing (HPC) Linux cluster resources, together with the minimum of management overhead.

View the White Paper Library

Multimedia

Video White Paper: Architecting a Better Network Storage Solution

BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.

Podcast: Interview with Ben Bennett of ClearSpeed Technology

Today, HPC organizations are requiring substantially more floating point performance to solve real-world problems. In this podcast, Ben Bennett, ClearSpeed General Manager, discusses how acceleration technology can improve the overall performance of standard x86-based systems...