Contact Us

|

Careers

|

Change Locale
close

Is Flex System your Shortcut to Convergence? [IBM]

Servers, Storage and Networking | Posted on April 25, 2013 by Doug Sekus

Unless your infrastructure is already standardized on IBM, choosing a solution like IBM Pure Systems may seem like a daunting, expensive rip-and-replace decisions. In a recent 8 page white paper, A Comparison of IBM’s New Flex System Environment to Traditional Blade Architectures, IT analyst firm Clabby Analytics compares IBM’s Flex System to its biggest rival – blade architecture.

In this post, the first in a 2-part series, we’ll examine their findings, introduce you to Flex System and answer the most frequently asked questions we hear from customers.

An introduction to Flex System
Flex System is a solution that provides users with the leeway to utilize non-IBM components when building their unique environments. Instead of having to cobble together a long list of servers, blades, storage arrays, network switches and adapters, Flex System delivers this via an all-in-one, pre-integrated chassis.

“Flex Systems are comprised of components that make up a whole system environment: compute nodes, storage nodes, network switches and PCIe expansion cards,” said Clabby Analytics.

What comes with Flex System?
Flex System compute nodes are available in standard rack and full-width form factors that are jam-packed with compute power:

  • The latest Intel Xeon x86 based processors (4-socket Xeon E5 multi-core processors)
  • 48 memory slots which can host up to 1.5 TB of memory
  • Two hot swappable 2.5 inch Solid State Drives (SSDs) or Hard Disk Drives (HDDs)
  • A LAN-on-Motherboard (LOM)

All of this is compressed into a chassis that takes up 10 Rack Units (RU) in a standard data center rack.

At first look, this may sound similar to existing blade server architectures. However, there are several distinct differences. According to the Clabby Analytics report:

“Flex System offers superior manageability, faster communications, more storage capacity (using up to eight internal SSDs per compute node) and better storage management – as well as broader/better physical/virtual system management – than all of today’s leading blade competitors.”

What management is like with Flex System
One of the benefits of Flex System is that it reduces management costs. Instead of relying on multiple applications from a variety of vendors, IBM’s cross-system management tool known as Flex System Manager (FSM) has the ability to manage across your entire environment from one screen. This means you manage all storage, networking, blade and communication components from one device.

These management capabilities even extend beyond Flex System itself and into other related equipment in the server room or data center. In the world of virtualization, Flex System is equipped with a dual enabled hypervisor as well as a VMware ESXi on flash key option. IBM’s VMControl module, for instance, can operate all major virtualization platforms from one interface.

Flex System Manager uses a ‘single-pane-of-glass’ management interface to manage across compute storage and networking resources. The ability to manage across an entire systems environment results in greater manager/administrator productivity — helping to significantly lower operational costs related to systems/storage/network management,” said Clabby Analytics.

How fast is it?
According to Clabby Analytics, one of the biggest differentiators of Flex System can be found in its network switching capabilities. Instead of a sending all traffic from the various compute nodes to the top of the rack and back which adds processing overhead and adds latency, Flex System’s internal switches cut latency in half. For organizations demanding high performance, this is a welcome advance.

The chassis holds up to 14 half-width compute nodes linked together with high-performance networking in the form of 10 Gb Ethernet switches with 40Gb uplinks, 16Gb Fibre Channel and the latest in InfiniBand connectivity.

How much storage capacity does it have?
When blade servers burst onto the scene a few years back, one of their major selling points was having little or no storage capacity on blade. This enabled designers to pack more processors and more memory into a smaller space. Unfortunately, the demands of a Big Data world mean that storage can no longer only be left to external storage arrays.

Accordingly, Flex System delivers up to eight internal SSDs per compute node for super-fast data access and up to 1.6 TB per node. Reliability features such as flash copy and thin provisioning have also been added. In the very near future, IBM will introduce a Storwise V7000-like storage node that can slot into the Flex System chassis if even more internal storage is desired.

“The first benefit of putting this large storage array within a Flex System rack is that compute nodes will need to take fewer hops to get to data. The second benefit is that the Storwize V7000 will be easier to manage,” said Clabby Analytics.

FSM simplifies how IT deals with pooling and virtualization of external storage, defining and attaching disks to virtual machines. Dynamic provisioning, storage tiering, and other advanced storage management functions can also be accomplished. For example, an IT manager working within FSM is able to launch IBM San Volume Controller to virtualize and manage all existing storage – regardless of vendor.

Do you think it’s too bleeding edge?
IBM designed the Flex System chassis to handle compute, storage, and networking requirements for the next decade. While individual components will be updated regularly, the chassis itself will not change. This protects the investment while enabling internal upgrades to be made at the owner’s discretion.

In the second blog post in this series, we’ll go into a detailed comparison of Flex System against the various blade server architectures currently on the market.

Related Articles

Cloud | August 19, 2019 by Softchoice Advisor

VMworld is the marquee VMware event of the year. The conference showcases the technology and solutions providers that are transforming the IT landscape. From mobility and the cloud to networking and security – VMworld offers a glimpse of what’s happening in IT now – and what’s coming next. The annual US conference kicks off in […]

Cloud | July 31, 2019 by Scott Mathewson

Most companies today use at least one cloud provider in some capacity.  Within two years, 92% of companies will be using two or more. This hybrid cloud world is forcing traditional data center design to evolve. The rise of hyper-converged systems and software-defined everything requires businesses to reevaluate traditional network and security designs to take […]

It’s fairly obvious that IBM’s POWER9 processors are, well, powerful. Staggeringly powerful. The evidence is straightforward: the Department of Energy’s Summit is the world’s fastest supercomputer, and it runs on POWER9 cores. Every day, Power Systems advance the frontiers of science on behalf of the world’s leading superpower. However, if you’re not the world’s leading […]