What’s the process to design an HP blade server?  Find out on my interview with Gary Thome, Chief Blade Architect for HP.  Thanks to the friends at SDR News, I was able to talk with Gary at the HP Tech Forum last month in Las Vegas.  In the video below, Gary discusses the steps it takes for HP to design the next generation blade and he discusses the latest blade servers announced at the HP Technology Forum.

Disclaimer: airfare, accommodations and some meals are being provided by HP, however the content being blogged is solely my opinion and does not in anyway express the opinions of HP.

Along with the Intel blade server announcements, Tuesday HP also announced two new AMD based blades, the BL465c G7 and BL685G7.  Although originally viewed as a refresh of their existing AMD blade servers, while at the HP Tech Forum in Las Vegas, I found there were a few interesting facts.  Let’s take a look.

BL465c G7
Before Tuesday’s announcement, HP had 2 Dual CPU AMD blades –  the BL495 blade server (based on the AMD Opteron 2300) and the BL465c G6 blade server (based on the AMD Opteron 2400.)  The BL495 had NO hot-plug drives, but maximum memory slots whereas the BL465c G6 offered fewer memory slots but offered hot-plug drives.  The 7th Generation (G7) of the BL465c takes a step toward combining the benefits of the BL495 and the Bl465c G6.  The BL465c G7 is a half-height blade server that has 2 AMD Opteron 6100 CPUs, supports up to 256GB of RAM AND offers 2 hot-plug drive bays. 

If you look at the image of the BL456c G7 on the left, you see only a single hot-plug drive – so where’s the 2nd?  It’s BEHIND the drive you see.  In a very innovative approach, the drives are stacked one behind the other but still “hot-plug” and serviceable without impacting the other drive.  It’s a bit hard to imagine, so take a few minutes to watch my video of the BL465c G7 from the HP Tech Forum EXPO floor (below.)  It’s also important to note this 7th generation server also comes standard with the dual port FlexFabric I/O standard on the motherboard as announced with the additional G7 blade models.

 Here are the specifics of the BL465c G7

• Up to two AMD Opteron™ 6100 Series processors
• Up to 256 GB of DDR3 memory
• 10GbE NC551i FlexFabric 2 Ports
• Smart Array P410i RAID controller
• Up to two hot plug disk drives
• Two mezzanine I/O expansion slots
• Insight Control with iLO Advanced (iLO 3)

For more information on the BL465c G7, check out HP’s web site.

BL685c G7
The 7th generation of the BL685c provides a 4 socket AMD Opteron 6100.  Code named, “Magny-Cours“, the AMD Opteron 6100 gives a substantial performance increase over the previous generation.  The BL685c G7 comes standard with 32 memory slots which can reach a max of 512GB of RAM. 

Here are the specifics of the BL685c G7:

• Up to four AMD Opteron™ 6100 Series processors
• Up to 512 GB of DDR3 memory
• (2) 10GbE NC551i FlexFabric 2 Ports per controller
• Smart Array P410i RAID controller
• Up to two hot plug disk drives
• Three mezzanine I/O expansion slots
• Insight Control with iLO Advanced (iLO 3)

For more information on the BL685c G7, check out HP’s web site.

Closer Look at the BL465c G7 and BL685c G7 from HP Tech Forum EXPO

At the 2010 HP Tech Forum press release on Monday, June 21, 2010, new HP BladeSystem products were announced. In the following video, Mark Potter, Vice President of HP Industry Standard Servers, reveals the newest HP BladeSystem products.

As expected, HP announced today new blade servers to their BladeSystem lineup as well as a new converged switch for their chassis. Everyone expected updates to the BL460 and BL490, the BL2x220c and even the BL680 blade servers, but the BL620 G7 blade server was a surprise (at least to me.)  Before I highlight the announcements, I have to preface this by saying I don’t have a lot of technical information yet.  I attended the press conference at HP Tech Forum 2010 in Las Vegas, but I didn’t get the press kit in advance.  I’ll update this post with links to the Spec Sheets as they become available.

The Details

First up- the BL620 G7

HP Proliant BL620 Blade Server

BL680 G7
The BL680 G7 is an upgrade to to the previous generation, however the 7th generation is a double-wide server.  This design offers up to 4 servers in a C7000 BladeSystem chassis.  This server’s claim to fame is that it will hold 1 terabyte (1TB) of RAM.  To put this into perspective, the Library of Congress’s entire library is 6TB of data, so you could put the entire library on 6 of these BL680 G7’s!  

HP BL680 G7 Blade Server

“FlexFabric” I/O Onboard
Each of the Generation 7 (G7) servers is coming with “FlexFabric” I/O on the motherboard of the blade server.  These are NEW NC551i Dual Port FlexFabric 10Gb Converged Network Adapters (CNA) that supports stateless TCP/IP offload, TCP Offload Engine (TOE), Fibre Channel over Ethernet (FCoE) and iSCSI protocols.  

Virtual Connect FlexFabric 10Gb/24-Port Module
The final “big” announcement on the blade server front is a converged fabric switch that fits inside the blade chassis.  Titled, the Virtual Connect FlexFabric 10Gb/24-Port Module, it is designed to allow you split the Ethernet fabric and the Fibre fabric at the switch module level, inside the blade chassis INSTEAD OF at the top of rack switch.  You may recall that I previously blogged about this as being a rumour, but now it is true.   

 The image on the left was my rendition of what it would look like.   

And here are the actual images.   

HP Virtual Connect FlexFabric 10Gb/24-Port Module

HP Virtual Connect FlexFabric 10Gb/24-Port Module

 HP believes that converged technology is good to the edge of the network, but that it is not yet mature enough to go to the datacenter (not ready for multi-hop, end to end.)  When the technology is acceptable and mature and business needs dictate – they’ll offer a converged network offering to the datacenter core.  

What do you think about these new blade offerings?  Let me know, I always enjoy your comments.  

Disclaimer: airfare, accommodations and some meals are being provided by HP, however the content being blogged is solely my opinion and does not in anyway express the opinions of HP.

It’s been a while since I’ve posted what rumours I’m hearing, so I thought I’d dig around and see what I can find out.  NOTE: this is purely speculation, I have no definitive information from any vendor about any of this information so this may be false info.  Read at your own risk.

Rumour #1 – GPU’s on a Blade Server
I’m hearing more and more discussion around “GPU’s” being used on a blade server.  Now, I have to admit, when I hear the term, “GPU“, I’m think of  Graphical Processing Unit – or the type of processor that runs a high-end graphics card.  So, when I hear rumours that there might be blade servers coming out that can handle GPUs, I have to wonder WHY? 

Wikipedia defines a GPU as “A graphics processing unit or GPU (also occasionally called visual processing unit or VPU) is a specialized processor that offloads 3D or 2D graphics rendering from the microprocessor. It is used in embedded systems, mobile phones, personal computers, workstations, and game consoles. Modern GPUs are very efficient at manipulating computer graphics, and their highly parallel structure makes them more effective than general-purpose CPUs for a range of complex algorithms. ” 

NVIDIA, the top maker of GPUs, also points out on their website, “The model for GPU computing is to use a CPU and GPU together in a heterogeneous computing model. The sequential part of the application runs on the CPU and the computationally-intensive part runs on the GPU. From the user’s perspective, the application just runs faster because it is using the high-performance of the GPU to boost performance. “ 

(For a cool Mythbusters video on GPU vs CPU, check out Cliff’s IBM Blog.)

So if a blade vendor decided to put together the ability to run normal AMD or Intel CPUs in tandem with GPU’s from NVIDIA, let’s say by using graphics cards in PCI-x expansion slots, they would have a blade server ideal for running any application that would benefit from high pherformace computing.  This seems do-able today since both HP and IBM offer PCI-x Expansion blades, however the rumour I’m hearing is that there is a blade server coming out that will be specifically designed for running GPUs.  Interesting concept.  I’m anxious to see how it will be received once it’s announced… 

Rumour #2 – Another Blade Server Dedicated for Memory
My second rumour is less exciting than the first – is that yet another blade vendor is about to announce a blade server designed for maximum memory density.  If you’ll recall, IBM has the HS22v blade and HP has the BL490c G6  blade server – both of which are designed for 18 memory DIMMs and internal drives.  So - that leaves either Cisco or Dell to be next on this rumour.  Since Cisco has the B250 blade server that can hold 48 DIMMs, I’m willing to believe they wouldn’t need to invest into designing a half-wide blade that can hold 18 DIMMs, therefore the only remaining option is Dell.  What would Dell gain from introducing a blade server with high memory density?  For one, it would give them an option to compete with IBM and HP in the “2 CPU, 18 Memory DIMM” environment.  Another reason is that it would help expand Dell’s blade portfolio.  If you examine what Dell’s current blade server offerings are today, you see they can’t compete with any requirement for large memory environments without moving to a full-height blade.

That’s all I have.  Let me know if you hear of any other rumours.

Chalk yet another win up for HP. 

It was reported last week on www.itnews.com.au that Digital production house Dr. D. Studios is in the early stages of building a supercomputer grid cluster for the rendering of the animated feature film Happy Feet 2 and visual effects in Fury Road the long-anticipated fourth film in the Mad Max series.  The super computer grid is based on HP BL490 G6 blade servers housed within an APC HACS pod, is already running in excess of 1000 cores and is expected to reach over 6000 cores during peak rendering by mid-2011.

This cluster boasted 4096 cores, taking it into the top 100 on the list of Top 500 supercomputers in the world in 2007 (it now sits at 447).

According to Doctor D infrastructure engineering manager James Bourne, “High density compute clusters provide an interesting engineering exercise for all parties involved. Over the last few years the drive to virtualise is causing data centres to move down a medium density path.”

Check out the full article, including video at:
http://www.itnews.com.au/News/169048,video-building-a-supercomputer-for-happy-feet-2-mad-max-4.aspx

Intel officially announced today the Xeon 5600 processor, code named “Westmere.” Cisco, HP and IBM also announced their blade servers that have the new processor. The Intel Xeon 5600 offers:

• 32nm process technology with 50% more threads and cache
• Improved energy efficiency with support for 1.35V low power memory

There will be 4 core and 6 core offerings. This processor also provide the option of HyperThreading, so you could have up to 8 threads and 12 threads per processor, or 16 and 24 in a dual CPU system. This will be a huge advantage to applications that like multiple threads, like virtualization. Here’s a look at what each vendor has come out with:

Cisco
The B200 M2 provides Cisco users with the current Xeon 5600 processors. It looks like Cisco will be offering a choice of the following Xeon 5600 processors: Intel Xeon X5670, X5650, E5640, E5620, L5640, or E5506. Because Cisco’s model is a “built-to-order” design, I can’t really provide any part numbers, but knowing what speeds they have should help.

HP
HP is starting off with the Intel Xeon 5600 by bumping their existing G6 models to include the Xeon 5600 processor. The look, feel, and options of the blade servers will remain the same – the only difference will be the new processor. According to HP, “the HP ProLiant G6 platform, based on Intel Xeon 5600 processors, includes the HP ProLiant BL280c, BL2x220c, BL460c and BL490c server blades and HP ProLiant WS460c G6 workstation blade for organizations requiring high density and performance in a compact form factor. The latest HP ProLiant G6 platforms will be available worldwide on March 29.” It appears that HP’s waiting until March 29 to provide details on their Westmere blade offerings, so don’t go looking for part numbers or pricing on their website.

IBM
IBM is continuing to stay ahead of the game with details about their product offerings. They’ve refreshed their HS22 and HS22v blade servers:

HS22
7870ECU – Express HS22, 2x Xeon 4C X5560 95W 2.80GHz/1333MHz/8MB L2, 4x2GB, O/Bay 2.5in SAS, SR MR10ie

7870G4U – HS22, Xeon 4C E5640 80W 2.66GHz/1066MHz/12MB, 3x2GB, O/Bay 2.5in SAS

7870GCU – HS22, Xeon 4C E5640 80W 2.66GHz/1066MHz/12MB, 3x2GB, O/Bay 2.5in SAS, Broadcom 10Gb Gen2 2-port

7870H2U -HS22, Xeon 6C X5650 95W 2.66GHz/1333MHz/12MB, 3x2GB, O/Bay 2.5in SAS

7870H4U – HS22, Xeon 6C X5670 95W 2.93GHz/1333MHz/12MB, 3x2GB, O/Bay 2.5in SAS

7870H5U – HS22, Xeon 4C X5667 95W 3.06GHz/1333MHz/12MB, 3x2GB, O/Bay 2.5in SAS

7870HAU – HS22, Xeon 6C X5650 95W 2.66GHz/1333MHz/12MB, 3x2GB, O/Bay 2.5in SAS, Emulex Virtual Fabric Adapter

7870N2U – HS22, Xeon 6C L5640 60W 2.26GHz/1333MHz/12MB, 3x2GB, O/Bay 2.5in SAS

7870EGU – Express HS22, 2x Xeon 4C E5630 80W 2.53GHz/1066MHz/12MB, 6x2GB, O/Bay 2.5in SAS

HS22V
7871G2U - HS22V, Xeon 4C E5620 80W 2.40GHz/1066MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871G4U - HS22V, Xeon 4C E5640 80W 2.66GHz/1066MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871GDU - HS22V, Xeon 4C E5640 80W 2.66GHz/1066MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871H4U - HS22V, Xeon 6C X5670 95W 2.93GHz/1333MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871H5U - HS22V, Xeon 4C X5667 95W 3.06GHz/1333MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871HAU - HS22V, Xeon 6C X5650 95W 2.66GHz/1333MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871N2U - HS22V, Xeon 6C L5640 60W 2.26GHz/1333MHz/12MB, 3x2GB, O/Bay 1.8in SAS

7871EGU - Express HS22V, 2x Xeon 4C E5640 80W 2.66GHz/1066MHz/12MB, 6x2GB, O/Bay 1.8in SAS

7871EHU - Express HS22V, 2x Xeon 6C X5660 95W 2.80GHz/1333MHz/12MB, 6x4GB, O/Bay 1.8in SAS

I could not find any information on what Dell will be offering, from a blade server perspective, so if you have information (that is not confidential) feel free send it my way.

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%

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."

Tolly.com announced on 2/25/2010 a new Test Report that compares the network bandwidth scalability between HP BladeSystem c7000 with BL460 G6 Servers and Cisco UCS 5100 with B200 Servers, and the results were interesting.   The report simply tested 6 HP blades, with a single Flex-10 Module vs 6 Cisco blades using their Fabric Extender + a single Fabric Interconnect.  I’m not going to try and re-state what the report says (for that you can download it directly), instead, I’m going to highlight the results.  It is important to note that the report was “commissioned by Hewlett-Packard Dev. Co, L.P.”

Result #1:  HP BladeSystem C7000 with a Flex-10 Module Tested to have More Aggregate Server Throughput (Gbps) over the Cisco UCS with a Fabric Extender connected to a Fabric Interconnect in a Physical-to-Physical Comparison
>The test shows when 4 physical servers were tested, Cisco can achieve an aggregate throughput of 36.59 Gbps vs HP achieving 35.83Gbps (WINNER: Cisco)

>When 6 physical servers were tested, Cisco achieved an aggregate throughput of 27.37 Gbps vs HP achieving 53.65 Gbps – a difference of 26.28 Gbps (WINNER: HP)

Result #2:  HP BladeSystem C7000 with a Flex-10 Module Tested to have More Aggregate Server Throughput (Gbps) over the Cisco UCS with a Fabric Extender connected to a Fabric Interconnect in a Virtual-to-Virtual Comparison
>Testing 2 servers each running 8 VMware Red Hat Linux hosts showed that HP achieved an aggregate throughput of 16.42 Gbps vs Cisco UCS achieving 16.70 Gbps (WINNER: Cisco). 

The results of the above was performed with the 2 x Cisco B200 blade servers each mapped to a dedicated 10Gb uplink port on the Fabric Extender (FEX).  When the 2 x Cisco B200 blade servers were designed to share the same 10Gb uplink port on the FEX, the achieved aggregate throughput on the Cisco UCS decreased to 9.10 Gbps.

A few points to note about these findings:
a) the HP Flex-10 Module has 8 x 10Gb uplinks whereas the Cisco Fabric Extender (FEX) has 4 x 10Gb uplinks

b) Cisco’s FEX Design allows for the 8 blade servers to extend out the 4 external ports in the FEX a 2:1 ratio (2 blades per external FEX port.) The current Cisco UCS design requires the servers to be “pinned”, or permanently assigned, to the respective FEX uplink. This works well when there are 4 blade servers, but when you get to more than 4 blade servers, the traffic is shared between two servers, which could cause bandwidth contention. 

 Furthermore, it’s important to understand that the design of the UCS blade infrastructure does not allow communication to go from Server 1 to Server 2 without leaving the FEX, connecting to the Fabric Interconnect (top of the picture) then returning to the FEX and connecting to the server.  This design is the potential cause of the decrease in aggregate throughput from 16.70Gbps to 9.10Gbps as shown above.

One of the “Bottom Line” conclusions from this report states, “throughput degradation on the Cisco UCS cased by bandwidth contention is a cause of concern for customers considering the use of UCS in a virtual server environment“  however I encourage you to take a few minutes, download this full report from the Tolly.com website and make your own conclusions about this report. 

Let me know your thoughts about this report – leave a comment below.

Disclaimer: This report was brought to my attention while attending the HP Tech Day event where airfare, accommodations and meals are being provided by HP, however the content being blogged is solely my opinion and does not in anyway express the opinions of HP.

Cisco is getting some (more) recognition with their UCS blade server product, as they recently achieved the top position for “8 Core Server” on VMware’s VMmark benchmark tool.  VMmark is the industry’s first (and only credible) virtualization benchmark for x86-based computers.  According to the VMmark website, the Cisco UCS B200 blade server reached a score of 25.06 @ 17 tiles.  A “tile” is simple a collection of virtual machines (VM’s) that are executing a set of diverse workloads designed to represent a natural work environment.   The total number of tiles that a server can handle provides a detailed measurement of that server’s consolidation capacity.

Cisco’s Winning Configuration
So – how did Cisco reach the top server spot?  Here’s the configuration:

server config:

• 2 x Intel Xeon X5570 Processors
• 96GB of RAM (16 x 8GB)
• 1 x Converged Network Adapter (Cisco UCS M71KR-Q)

storage config:

• EMC CX4-240
• Cisco MDS 9130
• 1154.27GB Used Disk Space
• 1024MB Array Cache
• 41 disks used on 4 enclosures/shelves (1 with 14 disk, 3 with 9 disks)
• 37 LUNs used
  *17 at 38GB (file server + mail server) over 20 x 73GB SSDs
  *17 at 15GB (database) + 2 LUNs at 400GB (Misc) over 16 x 450GB 15k disks
  * 1 LUN at 20GB (boot) over 5 x 300GB 15k disks
• RAID 0 for VMs, RAID 5 for VMware ESX 4.0 O/S

While first place on the VMmark page (8 cores) shows Fujitsu’s RX300, it’s important to note that it was reached using Intel’s W5590 processor – a processor that is designed for “workstations” – not servers.  Second place, of server processors, currently shows HP’s BL490 with 24.54 (@ 17 tiles)

Thanks to Omar Sultan (@omarsultan) for Tweeting about this and to Harris Sussman for blogging about it.