Posts

Fusion HCI Adds 8x the GPU Power

Image
IBM Storage Fusion HCI System has released a new GPU server today that takes a huge step forward in the support of AI applications.  The new server, called the G03, can be configured to have as many as 8 double-wide GPU adapter cards .   Fine-tuning and inferencing with foundation models used in AI applications increasingly requires faster and more powerful hardware, and the Fusion HCI G03 meets that need head on! The tremendous boost provided by the G03 for AI apps comes from having up to 8 of the  NVIDIA L40S  GPU PCIe double-wide adapter cards inside a single server.  The L40S is powered by the NVIDIA Ada Lovelace Architecture and is designed for generative AI and LLM training and inferencing.  It contains fourth generation tensor cores and the latest transformer engine that puts those tensor cores to good use. The design of the G03 provides flexibility for deploying GPU resources.  Many organizations will order the G03 with a full complement of 8 L40S GPUs.  But for organizations g
Post a Comment
Read more

Fusion HCI Performance Boost for AI Apps

Image
Fine-tuning and inferencing with foundation models used in AI applications increasingly requires faster and more powerful hardware as AI foundation models grow ever larger.  In response to this need, IBM Storage Fusion HCI System released a new GPU server option earlier this month called the G02 that gives a performance boost for AI applications. The new G02 GPU server is based on the Lenovo SR650 V3.  The configuration used in the G02 includes a pair of Intel Xeon Gold 6418H 24-core "Sapphire Rapids" processors.  This processor model includes Intel Deep Learning Boost, designed to accelerate AI use cases by extending the Intel AVX-512 instruction extension with a new instruction that increases deep learning inference performance. But the big boost provided by the G02 for AI apps comes from the three  NVIDIA A100 80GB GPU PCIe adapter cards in each G02 server.  This newer version of the A100 has u p to 1.25x higher AI inference performance over the A100 40GB  that was used i
Post a Comment
Read more

A New Direction for IBM Storage

Image
Yesterday, there were several significant announcements made about the branding and strategy of IBM's storage product portfolio.  And because the product that I so often write about is an important part of the portfolio, it was affected by these changes.   IBM has dropped the "Spectrum" brand and so what was IBM Spectrum Fusion HCI will now be known as IBM Storage Fusion HCI System . These changes are being made to align the portfolio with clients' most pressing data challenges.  IBM Storage solutions are now aligned under one of the three major data challenges faced today by organizations: AI and hyper data growth; connecting edge-to-core-to-cloud(s); and data resiliency.  Respectively, these challenges are addressed by the following IBM Storage solution categories: IBM Storage for Data and AI, IBM Storage for Hybrid Cloud, and IBM Storage for Data Resiliency. IBM Storage Fusion products are aligned under the category  IBM Storage for Hybrid Cloud . There was much mo
Post a Comment
Read more

Energizing IBM Spectrum Fusion HCI

Image
  Nothing happens in an IBM Spectrum Fusion HCI system without electricity.  I know that this is obvious, but sometimes we forget how important it is to get the supply of electrical power correct when setting up a system.  Fusion HCI is designed to be flexible when it comes to electrical power supply so that it can adapt to the local situation in countries all over the world.  Fusion HCI is also designed to be resilient to power supply failures.  Let's examine the components in Fusion HCI that contribute to that design. Layers of Power Resiliency Each Fusion HCI rack  contains multiple power distribution units (PDUs), and every component in the rack plugs into these PDUs.  In turn, each of these PDUs plugs into a power connection in the data center.  You can think of these PDUs as industrial-strength power bars.  Each PDU has twelve C13 receptacles into which the power cords from the rack components plug.  The C13 is the most common receptacle used by servers and switches a
Post a Comment
Read more

New Storage Server Options in Spectrum Fusion HCI 2.4

Image
On December 15, the new 2.4 version of IBM Spectrum Fusion HCI was released.  From a system hardware perspective, the most significant change in this release is the availability of powerful new storage servers.  The new storage servers have a total of 64 cores (128 threads) of processing power that comes from two AMD EPYC 3 7543 processors.  To support the applications using all these cores, there is 1024GB of RAM in each server, which works out to be 16GB/core.  While it's true that such 64-core servers were already available for Spectrum Fusion HCI, what is different now is that these servers come with a minimum of two Samsung 7.68TB PM1733 PCIe Gen 4 NVMe drives.  And they can be expanded to have up to 10 of these drives. With these new storage servers come new options for creating the base storage cluster in the system.  Previously, because there was only one storage server configuration available, there was only one way to build the base storage cluster and that wa
Post a Comment
Read more

Upgraded Compute Power for IBM Spectrum Fusion HCI

Image
  Big news!   A new and more powerful compute node is now available for the IBM Spectrum Fusion HCI product!  This new compute node provides double the number of cores and quadruple the amount of RAM compared to the original compute nodes that were released a year ago. CPU and Memory Like previous Spectrum Fusion HCI compute nodes, the new compute node is built upon a 1U Lenovo SR645 server.  The difference is that the two processor sockets inside are each populated with an AMD EPYC 7543 processor.  The EPYC 7543 is a 32-core, 64-thread processor that operates with a base clock speed of 2.8 GHz and comes with 256GB of cache.  With two of these processors, the server has a total of 64 cores and 128 threads available for running  applications on  Red Hat OpenShift. The AMD 7543 processor is part of the AMD EYPC 3 processor family, sometimes referred to as the "Milan" family.  AMD says that EPYC 3 brings a 19% improvement to its instructions per clock (IPC) performance  over
Post a Comment
Read more

Support for AI and ML Apps in IBM Spectrum Fusion HCI

Image
  In my last post , I went into detail describing the storage/compute servers that provide the storage cluster and most of the computing power in an IBM Spectrum Fusion HCI system .  In this post, I take a close look at one of the optional server types that can be added to a Spectrum Fusion HCI system, the GPU server .   It is an  NVIDIA-Certified System  that has passed certification tests to validate its performance, reliability, and scale for a wide range of workloads.   I detail the CPU, memory, storage, and network elements inside these GPU servers and, of course, the GPU adapters. CPU and Memory A Spectrum Fusion HCI GPU server is built from a Lenovo SR665 server that is two rack units (2U) high.  (A rack unit is an industry standard that translates to 1.75 inches; an industry standard data center rack used to hold servers and switches is 42U tall.)  Inside the server are two processor sockets that are each populated with an AMD EPYC 7F72 processor.  The EPYC 7F72 is a 24-core, 4
Post a Comment
Read more