IBM

You are currently browsing the archive for the IBM category.

4 Socket Blade Servers Density: Vendor Comparison

March 10, 2010 in Cisco, Dell, HP, IBM, Server Comparisons | View Comments

As the Intel Nehalem EX processor is a couple of weeks away, I wonder what impact it will have in the blade server market.  I’ve been talking about IBM’s HX5 blade server for several months now, so it is very clear that the blade server vendors will be developing blades that will have some iteration of the Xeon 7500 processor.  In fact, I’ve had several people confirm on Twitter that HP, Dell and even Cisco will be offering a 4 socket blade after Intel officially announces it on March 30.  For today’s post, I wanted to take a look at how the 4 socket blade space will impact the overall capacity of a blade server environment.  NOTE: this is purely speculation, I have no definitive information from any of these vendors that is not already public.

The Cisco UCS 5108 chassis holds 8 “half-width” B-200 blade servers or 4 “full-width” B-250 blade servers, so when we guess at what design Cisco will use for a 4 socket Intel Xeon 7500 (Nehalem EX) architecture, I have to place my bet on the full-width form factor.  Why?  Simply because there is more real estate.  The Cisco B250 M1 blade server is known for its large memory capacity, however Cisco could sacrifice some of that extra memory space for a 4 socket, “Cisco B350“ blade.  This would provide a bit of an issue for customers wanting to implement a complete rack full of these servers, as it would only allow for a total of 28 servers in a 42U rack (7 chassis x 4 servers per chassis.)

Estimated Cisco B300 with 4 CPUs

On the other hand, Cisco is in a unique position in that their half-width form factor also has extra real estate because they don’t have 2 daughter card slots like their competitors.  Perhaps Cisco would create a half-width blade with 4 CPUs (a B300?)  With a 42U rack, and using a half-width design, you would be able to get a maximum of 56 blade servers (7 chassis x 8 servers per chassis.)

Dell
The 10U M1000e chassis from Dell can currently handle 16 “half-height” blade servers or 8 “full height” blade servers.  I don’t forsee any way that Dell would be able to put 4 CPUs into a half-height blade.  There just isn’t enough room.  To do this, they would have to sacrifice something, like memory slots or a daughter card expansion slot, which just doesn’t seem like it is worth it.  Therefore, I predict that Dell’s 4 socket blade will be a full-height blade server, probably named a PowerEdge M910.  With this assumption, you would be able to get 32 blade servers in a 42u rack (4 chassis x 8 blades.) 

HP
Similar to Dell, HP’s 10U BladeSystem c7000 chassis can currently handle 16 “half-height” blade servers or 8 “full height” blade servers.  I don’t forsee any way that HP would be able to put 4 CPUs into a half-height blade.  There just isn’t enough room.  To do this, they would have to sacrifice something, like memory slots or a daughter card expansion slot, which just doesn’t seem like it is worth it.  Therefore, I predict that HP’s 4 socket blade will be a full-height blade server, probably named a Proliant BL680 G7 (yes, they’ll skip G6.)  With this assumption, you would be able to get 32 blade servers in a 42u rack (4 chassis x 8 blades.) 

IBM
Finally, IBM’s 9U BladeCenter H chassis offers up 14 servers.  IBM has one size server, called a “single wide.”  IBM will also have the ability to combine servers together to form a “double-wide”, which is what is needed for the newly announced IBM BladeCenter HX5.  A double-width blade server reduces the IBM BladeCenter’s capacity to 7 servers per chassis.  This means that you would be able to put 28 x 4 socket IBM HX5 blade servers into a 42u rack (4 chassis x 7 servers each.)

Summary
In a tie for 1st place, at 32 blade servers in a 42u rack, Dell and HP would have the most blade server density based on their existing full-height blade server design.  IBM and Cisco would come in at 3rd place with 28 blade servers in a 42u rack..  However IF Cisco (or HP and Dell for that matter) were able to magically re-design their half-height servers to hold 4 CPUs, then they would be able to take 1st place for blade density with 56 servers. 

Yes, I know that there are slim chances that anyone would fill up a rack with 4 socket servers, however I thought this would be good comparison to make.  What are your thoughts?  Let me know in the comments below.

Tags: , , , , , , , , , ,

IDC Q4 2009 Report: Blade Servers STILL Growing, HP Leads STILL Leading in Shares

March 8, 2010 in HP, IBM, IDC | View Comments

IDC reported on February 24, 2010 that blade server sales for Q4 2009 returned to quarterly revenue growth with factory revenues increasing 30.9% in Q4 2009 year over year (vs 1.2% in Q3.)  For the first time in 2009 there was an 8.3% increase in year-over-year shipments in Q4.  Overall blade servers accounted for $1.8 billion in Q4 2009 (up from $1.3 billion in Q3) which represented 13.9% of the overall server revenue.  It was also reported that more than 87% of all blade revenue in Q4 2009 was driven by x86 systems where blades now represent 21.4% of all x86 server revenue.

While the press release did not provide details of the market share for all of the top 5 blade vendors, they did provide data for the following: 

#1 market share: HP with 52.4%

#2 market share: IBM increased their marketshare from Q3 by 5.7% growth to 35.1%

q4_2009_idc

As an important note, according to IDC, IBM significantly outperformed the market with year-over-year revenue growth of 64.1%.  

According to Jed Scaramella, senior research analyst in IDC's Datacenter and Enterprise Server group,  "Blades remained a bright spot in the server vendors’ portfolios.  They were able to grow blade revenue throughout the year while maintaining their average selling prices. Customers recognize the benefits extend beyond consolidation and density, and are leveraging the platform to deliver a dynamic IT environment. Vendors consider blades strategic to their business due to the strong loyalty customers develop for their blade vendor as well as the higher level of pull-through revenue associated with blades."

Tags: , , , ,

Virtual I/O on IBM BladeCenter (IBM Virtual Fabric Adapter by Emulex)

March 4, 2010 in Emulex, IBM | View Comments

A few weeks ago, IBM and Emulex announced a new blade server adapter for the IBM BladeCenter and IBM System x line, called the “Emulex Virtual Fabric Adapter for IBM BladeCenter" (IBM part # 49Y4235). Frequent readers may recall that I had a "so what" attitude when I blogged about it in October and that was because, I didn't get it. I didn't get what the big deal was with being able to take a 10Gb pipe and allow you to carve it up into 4 "virtual NICs". HP's been doing this for a long time with their FlexNICs (check out VirtualKennth's blog for a great detail on this technology) so I didn't see the value in what IBM and Emulex was trying to do. But now I understand. Before I get into this, let me remind you of what this adapter is. The Emulex Virtual Fabric Adapter (CFFh) for IBM BladeCenter is a dual-port 10 Gb Ethernet card that supports 1 Gbps or 10 Gbps traffic, or up to eight virtual NIC devices.

This adapter hopes to address three key I/O issues:

1.Need for more than two ports per server, with 6-8 recommended for virtualization
2.Need for more than 1Gb bandwidth, but can't support full 10Gb today
3.Need to prepare for network convergence in the future

"1, 2, 3, 4"
I recently attended an IBM/Emulex partner event and Emulex presented a unique way to understand the value of the Emulex Virtual Fabric Adapter via the term, "1, 2, 3, 4" Let me explain:

"1" – Emulex uses a single chip architecture for these adapters. (As a non-I/O guy, I'm not sure of why this matters – I welcome your comments.)


"2" – Supports two platforms: rack and blade (Easy enough to understand, but this also emphasizes that a majority of the new IBM System x servers announced this week will have the Virtual Fabric Adapter "standard")

"3" – Emulex will have three product models for IBM (one for blade servers, one for the rack servers and one intergrated into the new eX5 servers)

"4" – There are four modes of operation:

  • Legacy 1Gb Ethernet
  • 10Gb Ethernet
  • Fibre Channel over Ethernet (FCoE)…via software entitlement ($$)
  • iSCSI Hardware Acceleration…via software entitlement ($$)

This last part is the key to the reason I think this product could be of substantial value. The adapter enables a user to begin with traditional Ethernet, then grow into 10Gb, FCoE or iSCSI without any physical change – all they need to do is buy a license (for the FCoE or iSCSI).

Modes of operation

The expansion card has two modes of operation: standard physical port mode (pNIC) and virtual NIC (vNIC) mode.

In vNIC mode, each physical port appears to the blade server as four virtual NIC with a default bandwidth of 2.5 Gbps per vNIC. Bandwidth for each vNIC can be configured from 100 Mbps to 10 Gbps, up to a maximum of 10 Gb per virtual port.

In pNIC mode, the expansion card can operate as a standard 10 Gbps or 1 Gbps 2-port Ethernet expansion card.

As previously mentioned, a future entitlement purchase will allow for up to two FCoE ports or two iSCSI ports. The FCoE and iSCSI ports can be used in combination with up to six Ethernet ports in vNIC mode, up to a maximum of eight total virtual ports.

Mode IBM Switch Compatibility

vNIC – works with BNT Virtual Fabric Switch
pNIC – works with BNT, IBM Pass-Thru, Cisco Nexus
FCoE- BNT or Cisco Nexus
iSCSI Acceleration – all IBM 10GbE switches

I really think the "one card can do all" concept works really well for the IBM BladeCenter design, and I think we'll start seeing more and more customers move toward this single card concept.

Comparison to HP Flex-10
I'll be the first to admit, I'm not a network or storage guy, so I'm not really qualified to compare this offering to HP's Flex-10, however IBM has created a very clever video that does some comparisons. Take a few minutes to watch and let me know your thoughts.

Tags: , , , ,

Announcing the IBM BladeCenter HX5 Blade Server (with detailed pics)

March 2, 2010 in IBM | View Comments

(UPDATED 11:29 AM EST 3/2/2010)
IBM announced today the BladeCenter® HX5 – their first 4 socket blade since the HS41 blade server.  IBM calls the HX5 “a scalable, high-performance blade server with unprecedented compute and memory performance, and flexibility ideal for compute and memory-intensive enterprise workloads.”   

The HX5 will have the ability to be coupled with a 2nd HX5 to scale to 4 CPU Sockets, grow beyond the base memory with the MAX5 memory expansion and be offer hardware partition to split a dual node server into 2 x single node servers and back again.  I’ll review each of these features in more detail below, but first, let’s look at the basics of the HX5 blade server.

X5 features:

  • Up to 2 x Intel Xeon 7500 CPUs per node
  • 16 DIMMs per node
  • 2 x Solid State Disk (SSD) slots per node
  • 1 x CIOv and 1 CFFh daughter card expansion slot per node, providing up to 8 I/O ports per node
  • 1 x scale connector per node

CPU Scalability
In the fashion of the eX5 architecture, IBM is enabling the HX5 blade server to grow from 2 CPUs to 4 CPUs (and theoretically more) via connecting the servers through a “scale connector“.  This connector will physically connect 2 HX5 servers on the tops of the servers, allowing the internal communications to extend to each others nodes.  The easiest way to think of this is like a Lego .  It will allow a HX5 or a MAX5 to be connected together.  There will be a 2 connector, a 3 connector and a 4 connector offering.  This means you could have any number of combinations from 2 x HX5 blade servers to 2 x HX5 blade servers + a MAX5 memory blade. 

Memory Scalability
With the addition of a new 24 DIMM memory blade, called the MAX5, IBM is enabling users to grow the base memory from 16 memory DIMMS to 48 40 (16+24) memory DIMMs. The MAX5 will be connected via the scale connector mentioned above, and in fact, when coupled with a 2 node, 4 socket system, could enable the entire system to have 72 80 DIMMS (16 DIMMs per HX5 plus 24 DIMMs per MAX5).  Granted, this will be a 4 server wide offering, but this will be a powerful offering for database servers, or even virtualization.

Hardware Partitioning
The final feature, known as FlexNode partitioning is the ability to split up a combined server node into individual server nodes and back again as needed.  Performed using IBM Software, this feature will enable a user to automatically take a 2 node HX5 system acting as a single 4 socket system and split it up into 2 x 2 socket systems then revert back to a single 4 socket system once the workload is completed. 

For example, during the day, the 4 socket HX5 server is used for as a database server, but at night, the database server is not being used, so the system is partitioned off into 2 x 2 socket physical servers that can each run their own applications. 

As I’ve mentioned previously, the pricing and part number info for the IBM BladeCenter HX5 blade server is not expected to show up until the Intel Xeon 7500 processor announcement on March 30, so when that info is released, you can find it here.

Let me know your thoughts – leave your comments below.

Tags: , , , ,

Announcing IBM eX5 Portfolio and the HX5 Blade Server

March 2, 2010 in IBM | View Comments

UPDATED: 3/2/2010 at 12:58 PM EST
Author’s Note: I’m stretching outside of my “blades” theme today so I can capture the entire eX5 messaging.
 
Finally, all the hype is over.  IBM announced today the next evolution of their “Enterprise x-Architecture”, also known as eX5.  
Why eX5?  Simple:  e=Enterprise X=x-Architecture  5=fifth generation. 

IBM’s Enterprise x-Architecture has been around for quite a while providing unique Scalability, Reliability and Flexibility in the x86 4-socket platforms.  You can check out the details of the eX4 technology here. 

Today’s announcement offered up a few facts:   

a) the existing x3850 and x3950 M2 will be called x3850 and x3950 X5 signifying a trend for IBM to move toward product naming designations that reflect the purpose of the server. 

b) the x3850 and x3950 X5’s will use the Intel Nehalem EX - to be officially announced/released on March 30.  At this time we can expect full details including part numbers, pricing and technical specifications. 

 c) a new 2u high,  2 socket server, the x3690 X5 was also announced.  This is probably the most exciting of the product announcements, as it is based on the Intel Nehalem EX processor but IBM’s innovation is going to enable the x3690 X5 to scale from 2 sockets to 4 sockets – but wait, there’s more.  There will be the ability, called MAX5 to add a memory expansion unit  to the x3690 X5 systems, enabling their system memory to be DOUBLED.d) in addition to the memory drawer, IBM will be shipping packs of solid state disks, called eXFlash that will deliver high performance to replace the limited IOPs of traditional spinning disks.  IBM is touting “significant” increases in performance for local databases with this new bundle of solid state disks.   In fact, according to IBM’s press release, eXFlash technology would eliminate the need for a client to purchase two entry-level servers and 80 JBODs to support a 240,000 IOPs database environment, saving $670,000 in server and storage acquisition costs.   The cool part is, these packs of disks will pop into the hot-swap drive bays of the x3690, x3850 and x3950 X5 servers.

e) IBM also announced a new technology, known as “FlexNode” that offers up physical partitioning capability for servers to move from being a single system to 2 different unique systems and back again. 

 
Blade Specific News
1) IBM will be releasing a new blade server, the BladeCenter HX5 next quarter that will also use the Intel Xeon 7500.  This blade server will scale, like all of the eX5 products, from 2 processors to 4 processors (and theoretically more) and will be ideal for database workloads.  Again, pricing and specs for this product will be released on the official Intel Nehalem EX launch date.  
 

 

  

IBM BladeCenter HX5 Blade Server

 

An observation from the pictures of the HX5 is that it will not have hot-swap drives, like the HS22’s do.  This means there will be internal drives – most like solid state drives (SSDs).  You may recall from my previous rumour post that the lack of hot-swap drives is pretty evident – IBM needed the real estate for the memory.  Unfortunately until memristors become available, blade vendors will need to sacrifice real estate for memory. 

2) As part of the MAX5 technology, IBM will also be launching a memory blade to increase the overall memory on the HX5 blade server.  Expect more details on this in the near future. 

Visit IBM’s website for their Live eX5 Event at 2 p.m. Eastern time at this site: 

http://www-03.ibm.com/systems/info/x86servers/ex5/events/index.html?CA=ex5launchteaser&ME=m&MET=exli&RE=ezvrm&Tactic=us0ab06w&cm_mmc=us0ab06w-_-m-_-ezvrm-_-ex5launchteaser-20100203 

As more information comes out on the new IBM eX5 portfolio, check back here and I’ll keep you posted.  I’d love to hear your thoughts in the comments below. 

MAX5 Memory Drawer (1U)

 

I find the x3690 X5 to be so interesting and exciting because it could quickly take over the server space that is currently occupied by the HP DL380 and the IBM x3650’s when it comes to virtualization.  We all know that VMware and other hypervisors thrive on memory, however the current 2 socket server design is limited to 12 – 16 memory sockets.  With the IBM System x3690 X5, this limitation can be overcome, as you can simply add on a memory drawer to achieve more memory capacity. 
Industry Opinions
Check out this analyst’s view of the IBM eX5 announcement here (pdf).
Here’s what VMware’s CTO, Stephen Herrod, has to say about IBM eX5:

  

Tags: , , , , , , , , , , , ,

IBM System x March 2 Event (What DOES 5 Mean?)

March 1, 2010 in IBM | View Comments

Tomorrow, March 2nd,  IBM kicks off a new portfolio of products in their System x line of product offerings.  One of the products will be a refresh and two of the products will be new.  However – don’t get your hopes up on seeing details on these new offerings, because tomorrow’s live event at 2 p.m. Eastern will be focused on the portfolio and the technology behind the portfolio.  IBM will not be disclosing any pricing, performance, model or Intel specifics until Intel’s launch dates on March 16 and March 30. 

“What Does 5 Mean to You” Campaign
5 business days ago, IBM kicked off a video campaign, “What Does 5 Mean to You.”  While a clever idea, I thought it missed on the messaging.  They were playing too much on “5″ – which will become clearer tomorrow when the announcement is made.  Here’s a look at all the videos:


What Are Your Top 5 IT Challenges (from the “What is 5″ videos)?
The key point of these videos were not to tease us, but to highlight the top 5 IT challenges that the new IBM portfolio will help solve.  Take a look at the top 5 IT challenges:

Challenge #5:  “My servers need Fibre Channel, Ethernet and iSCSI all operating at different speeds.  How do I simplify my networks right now?” 
Message:  Converged Infrastructure


Challenge #4:   “Why do I have to buy different types of servers whenever my needs change?  Can’t technology adapt to me?”
Message: Flexible Infrastructure

 

Challenge #3:  “My data costs keep growing.  How can I control the sprawl of my storage?”
Message: ?? Not Sure

Challenge #2:  “I don’t need a lot of complicated choices.  Why can’t I get a system that is set up for my workloads…right out of the box?”
Message: Like the IBM HS22v is “designed” for virtualization, we can expect this trend to continue with future IBM product releases…

Challenge #1:  “Technology competitors can all seem the same.  Doesn’t anyone have a game changing technology that will blow me away?”
Message: IBM expects the March 2nd announcement to be a game changer – and so do I. 

Check back with me tomorrow when IBM unveils What 5 Really Means!

Tags:

More HP and IBM Blade Rumours

February 24, 2010 in Cisco, HP, IBM | View Comments

I wanted to post a few more rumours before I head out to HP in Houston for “HP Blades and Infrastructure Software Tech Day 2010″ so it’s not to appear that I got the info from HP. NOTE: this is purely speculation, I have no definitive information from HP so this may be false info.

First off – the HP Rumour:
I’ve caught wind of a secret that may be truth, may be fiction, but I hope to find out for sure from the HP blade team in Houston.  The rumour is that HP’s development team currently has a Cisco Nexus Blade Switch Module for the HP BladeSystem in their lab, and they are currently testing it out.

Now, this seems far fetched, especially with the news of Cisco severing partner ties with HP, however, it seems that news tidbit was talking only about products sold with the HP label, but made by Cisco (OEM.)   HP will continue to sell Cisco Catalyst switches for the HP BladeSystem and even Cisco branded Nexus switches with HP part numbers (see this HP site for details.)  I have some doubt about this rumour of a Cisco Nexus Switch that would go inside the HP BladeSystem simply because I am 99% sure that HP is announcing a Flex10 type of BladeSystem switch that will allow converged traffic to be split out, with the Ethernet traffic going to the Ethernet fabric and the Fibre traffic going to the Fibre fabric (check out this rumour blog I posted a few days ago for details.)  Guess only time will tell.

The IBM Rumour:
I posted a few days ago a rumour blog that discusses the rumour of HP’s next generation adding Converged Network Adapters (CNA) to the motherboard on the blades (in lieu of the 1GB or Flex10 NICs), well, now I’ve uncovered a rumour that IBM is planning on following later this year with blades that will also have CNA’s on the motherboard.  This is huge!  Let me explain why. 

The design of IBM’s BladeCenter E and BladeCenter H have the 1Gb NICs onboard each blade server hard-wired to I/O Bays 1 and 2 – meaning only Ethernet modules can be used in these bays (see the image to the left for details.)  However, I/O Bays 1 and 2 are for ”standard form factor I/O modules” while I/O Bays are for “high speed form factor I/O modules”.  This means that I/O Bays 1 and 2 can not handle “high speed” traffic, i.e. converged traffic. 

 This means that IF IBM comes out with a blade server that has a CNA on the motherboard, either:

 a) the blade’s CNA will have to route to I/O Bays 7-10
OR
b) IBM’s going to have to come out with a new BladeCenter chassis that allows the high speed converged traffic from the CNAs to connect to a high speed switch module in Bays 1 and 2.

So let’s think about this.  If IBM (and HP for that matter) does put CNA’s on the motherboard, is there a need for additional mezzanine/daughter cards?  This means the blade servers could have more real estate for memory, or more processors.   If there’s no extra daughter cards, then there’s no need for additional I/O module bays.  This means the blade chassis could be smaller and use less power – something every customer would like to have.

I can really see the blade market moving toward this type of design (not surprising very similar to Cisco’s UCS design) – one where only a pair of redundant “modules” are needed to split converged traffic to their respective fabrics.  Maybe it’s all a pipe dream, but when it comes true in 18 months, you can say you heard it here first.

Thanks for reading.  Let me know your thoughts – leave your comments below.

Tags: , , , , , ,

Blade Networks Announces Industry’s First and Only Fully Integrated FCoE Solution Inside Blade Chassis

February 23, 2010 in BNT, IBM | View Comments

BLADE Network Technologies, Inc. (BLADE), ”officially” announces today the delivery of the industry’s first and only fully integrated Fibre Channel over Ethernet (FCoE) solution inside a blade chassis.   This integration significantly reduces power, cost, space and complexity over external FCoE implementations.

You may recall that I blogged about this the other day (click here to read), however I left off one bit of information.  The (Blade Networks) BNT Virtual Fabric 10 Gb Switch Module does not require the QLogic Virtual Fabric Extension Module to function.  It will work with an existing Top-of-Rack (TOR) Convergence Switch from Brocade or Cisco to act as a 10Gb switch module, feeding the converged 10Gb link up to the TOR switch.  Since it is a switch module, you can connect as few as 1 uplink to your TOR switch, therefore saving connectivity costs, as opposed to a pass-thru option (click here for details on the pass-thru option.) 

Yes – this is the same architectural design as the Cisco Nexus 4001i provides as well, however there are a couple of differences:

BNT Virtual Fabric Switch Module (IBM part #46C7191) - 10 x 10Gb Uplinks, $11,199 list (U.S.)
Cisco Nexus 4001i Switch (IBM part #46M6071) - 6 x 10Gb Uplinks, $12,999 list (U.S.)

While BNT provides 4 extra 10Gb uplinks, I can’t really picture anyone using all 10 ports.  However, it does has a lower list price, but I encourage you to check your actual price with your IBM partner, as the actual pricing may be different.  Regardless of whether you choose BNT or Cisco to connect into your TOR switch, don’t forget the transceivers!  They add much more $$ to the overall cost, and without them you are hosed.

About the BNT Virtual Fabric 10Gb Switch Module
The BNT Virtual Fabric 10Gb Switch Module includes the following features and functions:

  • Form-factor
    • Single-wide high-speed switch module (fits in IBM BladeCenter H bays #7 and 9.) 
  • Internal ports
    • 14 internal auto-negotiating ports: 1 Gb or 10 Gb to the server blades
    • Two internal full-duplex 100 Mbps ports connected to the management module
  • External ports
    • Up to ten 10 Gb SFP+ ports (also designed to support 1 Gb SFP if required, flexibility of mixing 1 Gb/10 Gb)
    • One 10/100/1000 Mb copper RJ-45 used for management or data
    • An RS-232 mini-USB connector for serial port that provides an additional means to install software and configure the switch module
  • Scalability and performance
    • Autosensing 1 Gb/10 Gb internal and external Ethernet ports for bandwidth optimization

To read the extensive list of details about this switch, please visit the IBM Redbook located here.

Tags: , , , , , ,

IBM’s New Approach to Ethernet/Fibre Traffic

February 18, 2010 in IBM | View Comments

Okay, I’ll be the first to admit when I’m wrong – or when I provide wrong information. 

 A few days ago, I commented that no one has yet offered the ability to split out Ethernet and Fibre traffic at the chassis level (as opposed to using a top of rack switch.)  I quickly found out that I was wrong – IBM now has the ability to separate the Ethernet fabric and the Fibre fabric at the BladeCenter H,  so if you are interested grab a cup of coffee and enjoy this read. 

First a bit of background.  The traditional method of providing Ethernet and Fibre I/O in a blade infrastructure was to integrate 6 Ethernet switches and 2 Fibre switches into the blade chassis, which provides 6 NICs and 2 Fibre HBAs per blade server.  This is a costly method and it limits the scalability of a blade server. 

A more conventional method that is becoming more popular is to converge the I/O traffic using a single converged network adapter (CNA) to carry the Ethernet and the Fibre traffic over a single 10Gb connection to a top of rack (TOR) switch which then sends the Ethernet traffic to the Ethernet fabric and the Fibre traffic to the Fibre fabric.  This reduces the number of physical cables coming out of the blade chassis, offers higher bandwidth and reduces the overall switching costs.  Up now, IBM offered two different methods to enable converged traffic:

Method 1: connect a pair of 10Gb Ethernet Pass-Thru modules into the blade chassis, add a CNA on each blade server, then connect the pass thru modules to a top of rack  convergence switch from Brocade or Cisco.  This method is the least expensive method, however since Pass-Thru modules are being used, a connection is required on the TOR convergence switch for every blade server being connected.  This would mean a 14 blade infrastructure would eat up 14 ports on the convergence switch, potentially leaving the switch with very few available ports.

Method #2: connect a pair of IBM Cisco Nexus 4001i switches, add a CNA on each server then connect the Nexus 4001i to a Cisco Nexus 5000 top of rack switch.  This method enables you to use as few as 1 uplink connection from the blade chassis to the Nexus 5000 top of rack switch, however it is more costly and you have to invest into another Cisco switch.

The New Approach
A few weeks ago, IBM announced the “Qlogic Virtual Fabric Extension Module” – a device that fits into the IBM BladeCenter H and takes the the Fibre traffic from the CNA on a blade server and sends it to the Fibre fabric.  This is HUGE!  While having a top of rack convergence switch is helpful, you can now remove the need to have a top of rack switch because the I/O traffic is being split out into it’s respective fabrics at the BladeCenter H. 

What’s Needed
I’ll make it simple – here’s a list of components that are needed to make this method work:

  • 2 x BNT Virtual Fabric 10 Gb Switch Module – part # 46C7191
  • 2 x QLogic Virtual Fabric Extension Module – part # 46M6172
  • a Qlogic 2-port 10 Gb Converged Network Adapter per blade server – part # 42C1830
  • a IBM 8 Gb SFP+ SW Optical Transceiver for each uplink needed to your fibre fabric – part # 44X1964 (note – the QLogic Virtual Fabric Extension Module doesn’t come with any, so you’ll need the same quantity for each module.)

The CNA cards connect to the BNT Virtual Fabric 10 Gb Switch Module in Bays 7 and 9.  These switch modules have an internal connector to the QLogic Virtual Fabric Extension Module, located in Bays 3 and 5.  The I/O traffic moves from the CNA cards to the BNT switch, which separates the Ethernet traffic and sends it out to the Ethernet fabric while the Fibre traffic routes internally to the QLogic Virtual Fabric Extension Modules.  From the Extension Modules, the traffic flows into the Fibre Fabric.

It’s important to understand the switches, and how they are connected, too, as this is a new approach for IBM.  Previously the Bridge Bays (I/O Bays 5 and 6) really haven’t been used and IBM has never allowed for a card in the CFF-h slot to connect to the switch bay in I/O Bay 3. 

 

There are a few other designs that are acceptable that will still give you the split fabric out of the chassis, however they were not “redundant” so I did not think they were relevant.  If you want to read the full IBM Redbook on this offering, head over to IBM’s site.

A few things to note with the maximum redundancy design I mentioned above:

1) the CIOv slots on the HS22 and HS22v can not be used.  This is because I/O bay 3 is being used for the Extension Module and since the CIOv slot is hard wired to I/O bay 3 and 4, that will just cause problems – so don’t do it.

2) The BladeCenter E chassis is not supported for this configuration.  It doesn’t have any “high speed bays” and quite frankly wasn’t designed to handle high I/O throughput like the BladeCenter H.

3) Only the parts listed above are supported.  Don’t try and slip in a Cisco Fibre Switch Module or use the Emulex Virtual Adapter on the blade server – it won’t work.  This is a QLogic design and they don’t want anyone else’s toys in their backyard.

That’s it.  Let me know what you think by leaving a comment below.  Thanks for stopping by!

Tags: , , , , ,

Introducing the IBM HS22v Blade Server

February 9, 2010 in IBM | View Comments

IBM officially announced today a new addition to their blade server line – the HS22v.  Modeled after the HS22 blade server, the HS22v is touted by IBM as a “high density, high performance blade optimized for virtualization.”  So what makes it so great for virtualization?  Let’s take a look.

Memory
One of the big differences between the HS22v and the HS22 is more memory slots.  The HS22v comes with 18 x very low profile (VLP) DDR3 memory DIMMs for a maximum of 144GB RAM.  This is a key attribute for a server running virtualization since everyone knows that VM’s love memory.  It is important to note, though, the memory will only run at 800Mhz when all 18 slots are used.  In comparison, if you only had 6 memory DIMMs installed (3 per processor) then the memory would run at 1333Mhz and 12 DIMMs installed (6 per processor) runs at 1066Mhz.  As a final note on the memory, this server will be able to use both 1.5v and 1.35v memory.  The 1.35v will be newer memory that is introduced as the Intel Westmere EP processor becomes available.  The big deal about this is that lower voltage memory = lower overall power requirements.

Drives
The second big difference is the HS22v does not use hot-swap drives like the HS22 does.  Instead, it uses a 2 x solid state drives (SSD) for local storage. These drives have  hardware RAID 0/1 capabilities standard.  Although the picture to the right shows a 64GB SSD drive, my understanding is that only 50GB drives will be available as they start to become readlily available on March 19, with larger sizes (64GB and 128GB) becoming available in the near future.  Another thing to note is that the image shows a single SSD drive, however the 2nd drive is located directly beneath.  As mentioned above, these drives do have the ability to be set up in a RAID 0 or 1 as needed.

So – why did IBM go back to using internal drives?  For a few reasons:

Reason #1 : in order to get the space to add the extra memory slots, a change had to be made in the design.  IBM decided that solid state drives were the best fit.

Reason #2: the SSD design allows the server to run with lower power.  It’s well known that SSD drives run at a much lower power draw than physical spinning disks, so using SSD’s will help the HS22v be a more power efficient blade server than the HS22.

Reason #3: a common trend of virtualization hosts, especially VMware ESXi, is to run on integrated USB devices.  By using an integrated USB key for your virtualization software, you can eliminate the need for spinning disks, or even SSD’s therefore reducing your overall cost of the server.

Processors
So here’s the sticky area.  IBM will be releasing the HS22v with the Intel Xeon 5500 processor first.  Later in March, as the Intel Westmere EP (Intel Xeon 5600) is announced, IBM will have models that come with it.  IBM will have both Xeon 5500 and Xeon 5600 processor offerings.  Why is this?  I think for a couple of reasons:

a) the Xeon 5500 and the Xeon 5600 will use the same chipset (motherboard) so it will be easy for IBM to make one server board, and plop in either the Nehalem EP or the Westmere EP

b) simple – IBM wants to get this product into the marketplace sooner than later.

Questions

1) Will it fit into the BladeCenter E? 
YES – however there may be certain limitations, so I’d recommend you reference the IBM BladeCenter Interoperability Guide for details.

2) Is it certified to run VMware ESX 4?
YES

3) Why didn’t IBM call it HS22XM?
According to IBM, the “XM” name is feature focused while “V” is workload focused - a marketing strategy we’ll probably see more of from IBM in the future.

That’s it for now.  If there are any questions you have about the HS22v, let me know in the comments and I’ll try to get some answers.

For more on the IBM HS22v, check out IBM’s web site here.

Check back with me in a few weeks when I’m able to give some more info on what’s coming from IBM!

Tags: , , , , ,

« Older entries