Connecting the Dots: Applications and Grid Infrastructure

By By Yaron Haviv, CTO, Voltaire

May 28, 2007

The move to grid or grid-like architectures within the datacenter brings many benefits, such as growth of capacities, lower costs, and support for an increasing number and variety of applications. This trend has also brought additional infrastructure requirements and associated challenges. Connecting and managing hundreds or thousands of servers and networked storage, and incorporating server and storage virtualization technologies, has created communication challenges, network complexity and a steep learning curve for getting the most out of the ability to virtualize infrastructure.

With all of this complexity to deal with, have we figured out how much time and resources it takes to deploy applications over grids?

This article examines how grids built around a service-oriented architecture (SOA) focusing on business tasks, business flows, and service delivery will significantly shorten the time and efforts in application deployment and configuration, while delivering the greatest efficiencies. A datacenter grid model is proposed, which includes considerations for deployment and provisioning tools for applications, server and storage infrastructure, and high-performance grid fabrics.

The Evolution of Today’s Data Center Challenges

In recent years, IT has been faced with enormous growth in capacities. Digital data is generated at a growing rate and covers every aspects of our lives — from telephones (VoIP, cellular) to media and entertainment, retail, banking and leisure. The amount of data grows exponentially and becomes larger and more complex to process (think about types, such as media, XML, DNA, etc.), requiring faster servers and storage with greater capacities.

Because compute and storage capacities needed to grow to address the data influx and budgets remained flat, these trends forced a paradigm shift and led to increased commoditization of the compute and storage infrastructure. Commodity-based server clusters and grids replaced large and expansive mainframes, and clusters of low- to mid-range storage replaced the “Big Irons.” Furthermore, server and storage virtualization technologies were introduced to increase the utilization of the hardware, leading to hardware cost savings.

While this revolution in server and storage architecture saves significant amounts of money on hardware and capital expenses, it also created new and increased software and operations expenses. IT’s next challenge is to increase the software efficiency and to reduce the operational costs in grids.

The Key to Reducing Software and Operational Costs? Simplify the Datacenter!

The need for increased capacities at lower costs drove the industry to focus on reducing capital expenses and hardware costs. While doing this, the datacenter became more complex and fragmented, driving up operational costs. Unary systems are replaced with server farms and clusters, network topologies become more complicated due to the distributed nature of the new data center, and new technologies such as SAN (storage area networks) introduced with them complete processes and disciplines (not to mention additional organizational barriers). Server virtualization further fragmented the resources, creating more manageable components and undermining some key assumptions about the relationship between a resource and its physical implementation. Provisioning of an application environment that consists of several tiers can take weeks and relies on many resources. Even small changes to the environment may require scheduling downtime and serial processes that take days.

To address complexity, datacenters will need to be built around an SOA, which focuses on business tasks, business flows and service delivery. This new architecture transcends hardware and software. The infrastructure will become adaptive and self-configuring, focusing less on technology silos and manual processes that take much time and resources. It will instead zero in on how the infrastructure can satisfy the service and application objectives. This “subtlety” is a prerequisite to constructing a true SOA.

Simplification Through I/O and Fabric Consolidation

A significant barrier to achieving a utility datacenter is the tight relation that exists between applications and infrastructure. Some applications are more dependent on compute, some on storage, and some rely more heavily on low-latency, inter-processor communication. Each application may drive different hardware requirements or configurations. For example, a standard Web server and a standard 1U server with two network ports may be good enough for some applications, but for a database of file servers, more and faster I/O is needed. Without addressing that relationship, datacenters would still need to be provisioned physically or, in the best case, automation and virtualization tools would be limited to homogeneous silos.

And CPU capacities are growing. The world is migrating to 64-bit CPU technologies from Intel and AMD, CPUs are faster, and multi-core technology with dual- or quad-CPU cores on a single chip is now reality. New power-saving and cooling technologies allow higher system densities increasing the overall CPU capacity per server by a factor of 10 or more. This requires an equivalent capacity growth on the server I/O, storage and network interfaces.

To address these I/O challenges, servers now need 10 Gbps or faster external interfaces that can process and/or virtualize I/O in hardware. Middleware is needed that can take advantage of that hardware and bypass the OS while still providing traditional application and/or storage APIs. Moreover, the communication fabric/network should be more reliable and predictable — and not waste precious application run time on wait or retransmissions. Since network, storage and I/O resources are scarce, it also is important to enable efficient management and dynamically partition the I/O resources to the right applications and traffic. Much like the CPU-partitioning executed by server virtualization technologies, this can also eliminate the need for manual infrastructure and cable provisioning, and shorten application deployment time.

In today’s datacenter, a lot of focus in put on incorporating tools for storage and server automation, which can take care of many deployment and maintenance tasks. However, a key element in the creation of an operationally efficient datacenter is the use of a unified fabric (for all cluster, I/O and network traffic) that can allow for partitioning of the fabric and the attached I/O to accommodate dynamic application needs. The fabric consolidation is achieved using multi-service switches and servers are connected to the fabric switch using a multi-channel adapter (such as an InfiniBand adapter). This configuration can provide multiple virtual NICs, fast access to storage and low-latency messaging and RDMA for application scale-out and clustering. The switches can form multiple virtual and isolated LAN, SAN and cluster networks on-demand, and attach transparently to external Ethernet or Fibre Channel networks.

Using InfiniBand fabric technology, a single adapter and single link can emulate multiple adapters and network ports. InfiniBand has built-in mechanisms to ensure isolation between virtual I/O elements and between different logical networks. 10 GbE can also be used as a lower-end alternative to InfiniBand. However, today, 10 GbE’s pricing remains high and the technology is immature by comparison. Furthermore, InfiniBand has many unique capabilities that enable consolidation that are not found in 10 GbE.

Service-Oriented Infrastructure (SOI) Management

While virtualization of servers, storage and fabrics is a key element to achieving a flexible and more efficient datacenter, it also is critical to develop a new approach to data center resource management. Instead of manual procedures by which administrators create and configure the infrastructure, infrastructure resources should be dynamically created and configured based on the application requirements. This is achieved through the use of SOI management tools.

Fabric provisioning and SOI management tools, such as Voltaire GridVision Enterprise software, depend on the use of dynamic and unified datacenter fabrics, which have loose relationships between resources and can be programmed to create whatever topology or logical links are needed at a given time or to satisfy a given application load.

These tools are complementary to many of the virtualization and automation/provisioning tools in market today because they focus on the infrastructure and connectivity aspects of virtual datacenter resources. They can integrate with the server virtualization products (such as Xen and VMWare) and typically use an open and extensible API for optional integration with server and storage provisioning tools  Orchestration and scheduling tools can use the SOI Web services API and object models to provision infrastructure as needed, collect health/performance information and get notified on infrastructure events and changes.

With a SOI, equipment can be wired once, thus eliminating physical user intervention. Complex application deployment procedures that cross organizational boundaries can be automated and conducted in few minutes rather than days or weeks. They are less error-prone and consume fewer resources. Furthermore, infrastructure can be built-to-order to meet application-specific requirements with the right balance of CPU, network and storage resources. Ultimately, this makes applications on a grid more efficient and eliminates the right bottleneck.

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!

Mira Supercomputer Enables Cancer Research Breakthrough

November 11, 2019

Dynamic partial-wave spectroscopic (PWS) microscopy allows researchers to observe intracellular structures as small as 20 nanometers – smaller than those visible by optical microscopes – in three dimensions at a mill Read more…

By Staff report

IBM Adds Support for Ion Trap Quantum Technology to Qiskit

November 11, 2019

After years of percolating in the shadow of quantum computing research based on superconducting semiconductors – think IBM, Rigetti, Google, and D-Wave (quantum annealing) – ion trap technology is edging into the QC Read more…

By John Russell

Tackling HPC’s Memory and I/O Bottlenecks with On-Node, Non-Volatile RAM

November 8, 2019

On-node, non-volatile memory (NVRAM) is a game-changing technology that can remove many I/O and memory bottlenecks and provide a key enabler for exascale. That’s the conclusion drawn by the scientists and researcher Read more…

By Jan Rowell

What’s New in HPC Research: Cosmic Magnetism, Cryptanalysis, Car Navigation & More

November 8, 2019

In this bimonthly feature, HPCwire highlights newly published research in the high-performance computing community and related domains. From parallel programming to exascale to quantum computing, the details are here. Read more…

By Oliver Peckham

Machine Learning Fuels a Booming HPC Market

November 7, 2019

Enterprise infrastructure investments for training machine learning models have grown more than 50 percent annually over the past two years, and are expected to shortly surpass $10 billion, according to a new market fore Read more…

By George Leopold

AWS Solution Channel

Making High Performance Computing Affordable and Accessible for Small and Medium Businesses with HPC on AWS

High performance computing (HPC) brings a powerful set of tools to a broad range of industries, helping to drive innovation and boost revenue in finance, genomics, oil and gas extraction, and other fields. Read more…

IBM Accelerated Insights

Atom by Atom, Supercomputers Shed Light on Alloys

November 7, 2019

Alloys are at the heart of human civilization, but developing alloys in the Information Age is much different than it was in the Bronze Age. Trial-by-error smelting has given way to the use of high-performance computing Read more…

By Oliver Peckham

IBM Adds Support for Ion Trap Quantum Technology to Qiskit

November 11, 2019

After years of percolating in the shadow of quantum computing research based on superconducting semiconductors – think IBM, Rigetti, Google, and D-Wave (quant Read more…

By John Russell

Tackling HPC’s Memory and I/O Bottlenecks with On-Node, Non-Volatile RAM

November 8, 2019

On-node, non-volatile memory (NVRAM) is a game-changing technology that can remove many I/O and memory bottlenecks and provide a key enabler for exascale. Th Read more…

By Jan Rowell

MLPerf Releases First Inference Benchmark Results; Nvidia Touts its Showing

November 6, 2019

MLPerf.org, the young AI-benchmarking consortium, today issued the first round of results for its inference test suite. Among organizations with submissions wer Read more…

By John Russell

Azure Cloud First with AMD Epyc Rome Processors

November 6, 2019

At Ignite 2019 this week, Microsoft's Azure cloud team and AMD announced an expansion of their partnership that began in 2017 when Azure debuted Epyc-backed ins Read more…

By Tiffany Trader

Nvidia Launches Credit Card-Sized 21 TOPS Jetson System for Edge Devices

November 6, 2019

Nvidia has launched a new addition to its Jetson product line: a credit card-sized (70x45mm) form factor delivering up to 21 trillion operations/second (TOPS) o Read more…

By Doug Black

In Memoriam: Steve Tuecke, Globus Co-founder

November 4, 2019

HPCwire is deeply saddened to report that Steve Tuecke, longtime scientist at Argonne National Lab and University of Chicago, has passed away at age 52. Tuecke Read more…

By Tiffany Trader

Spending Spree: Hyperscalers Bought $57B of IT in 2018, $10B+ by Google – But Is Cloud on Horizon?

October 31, 2019

Hyperscalers are the masters of the IT universe, gravitational centers of increasing pull in the emerging age of data-driven compute and AI.  In the high-stake Read more…

By Doug Black

Cray Debuts ClusterStor E1000 Finishing Remake of Portfolio for ‘Exascale Era’

October 30, 2019

Cray, now owned by HPE, today introduced the ClusterStor E1000 storage platform, which leverages Cray software and mixes hard disk drives (HDD) and flash memory Read more…

By John Russell

Supercomputer-Powered AI Tackles a Key Fusion Energy Challenge

August 7, 2019

Fusion energy is the Holy Grail of the energy world: low-radioactivity, low-waste, zero-carbon, high-output nuclear power that can run on hydrogen or lithium. T Read more…

By Oliver Peckham

Using AI to Solve One of the Most Prevailing Problems in CFD

October 17, 2019

How can artificial intelligence (AI) and high-performance computing (HPC) solve mesh generation, one of the most commonly referenced problems in computational engineering? A new study has set out to answer this question and create an industry-first AI-mesh application... Read more…

By James Sharpe

Cray Wins NNSA-Livermore ‘El Capitan’ Exascale Contract

August 13, 2019

Cray has won the bid to build the first exascale supercomputer for the National Nuclear Security Administration (NNSA) and Lawrence Livermore National Laborator Read more…

By Tiffany Trader

DARPA Looks to Propel Parallelism

September 4, 2019

As Moore’s law runs out of steam, new programming approaches are being pursued with the goal of greater hardware performance with less coding. The Defense Advanced Projects Research Agency is launching a new programming effort aimed at leveraging the benefits of massive distributed parallelism with less sweat. Read more…

By George Leopold

AMD Launches Epyc Rome, First 7nm CPU

August 8, 2019

From a gala event at the Palace of Fine Arts in San Francisco yesterday (Aug. 7), AMD launched its second-generation Epyc Rome x86 chips, based on its 7nm proce Read more…

By Tiffany Trader

D-Wave’s Path to 5000 Qubits; Google’s Quantum Supremacy Claim

September 24, 2019

On the heels of IBM’s quantum news last week come two more quantum items. D-Wave Systems today announced the name of its forthcoming 5000-qubit system, Advantage (yes the name choice isn’t serendipity), at its user conference being held this week in Newport, RI. Read more…

By John Russell

Ayar Labs to Demo Photonics Chiplet in FPGA Package at Hot Chips

August 19, 2019

Silicon startup Ayar Labs continues to gain momentum with its DARPA-backed optical chiplet technology that puts advanced electronics and optics on the same chip Read more…

By Tiffany Trader

Crystal Ball Gazing: IBM’s Vision for the Future of Computing

October 14, 2019

Dario Gil, IBM’s relatively new director of research, painted a intriguing portrait of the future of computing along with a rough idea of how IBM thinks we’ Read more…

By John Russell

Leading Solution Providers

ISC 2019 Virtual Booth Video Tour

CRAY
CRAY
DDN
DDN
DELL EMC
DELL EMC
GOOGLE
GOOGLE
ONE STOP SYSTEMS
ONE STOP SYSTEMS
PANASAS
PANASAS
VERNE GLOBAL
VERNE GLOBAL

Intel Confirms Retreat on Omni-Path

August 1, 2019

Intel Corp.’s plans to make a big splash in the network fabric market for linking HPC and other workloads has apparently belly-flopped. The chipmaker confirmed to us the outlines of an earlier report by the website CRN that it has jettisoned plans for a second-generation version of its Omni-Path interconnect... Read more…

By Staff report

Kubernetes, Containers and HPC

September 19, 2019

Software containers and Kubernetes are important tools for building, deploying, running and managing modern enterprise applications at scale and delivering enterprise software faster and more reliably to the end user — while using resources more efficiently and reducing costs. Read more…

By Daniel Gruber, Burak Yenier and Wolfgang Gentzsch, UberCloud

Dell Ramps Up HPC Testing of AMD Rome Processors

October 21, 2019

Dell Technologies is wading deeper into the AMD-based systems market with a growing evaluation program for the latest Epyc (Rome) microprocessors from AMD. In a Read more…

By John Russell

Intel Debuts Pohoiki Beach, Its 8M Neuron Neuromorphic Development System

July 17, 2019

Neuromorphic computing has received less fanfare of late than quantum computing whose mystery has captured public attention and which seems to have generated mo Read more…

By John Russell

Rise of NIH’s Biowulf Mirrors the Rise of Computational Biology

July 29, 2019

The story of NIH’s supercomputer Biowulf is fascinating, important, and in many ways representative of the transformation of life sciences and biomedical res Read more…

By John Russell

Xilinx vs. Intel: FPGA Market Leaders Launch Server Accelerator Cards

August 6, 2019

The two FPGA market leaders, Intel and Xilinx, both announced new accelerator cards this week designed to handle specialized, compute-intensive workloads and un Read more…

By Doug Black

With the Help of HPC, Astronomers Prepare to Deflect a Real Asteroid

September 26, 2019

For years, NASA has been running simulations of asteroid impacts to understand the risks (and likelihoods) of asteroids colliding with Earth. Now, NASA and the European Space Agency (ESA) are preparing for the next, crucial step in planetary defense against asteroid impacts: physically deflecting a real asteroid. Read more…

By Oliver Peckham

When Dense Matrix Representations Beat Sparse

September 9, 2019

In our world filled with unintended consequences, it turns out that saving memory space to help deal with GPU limitations, knowing it introduces performance pen Read more…

By James Reinders

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