One of the industry-leading independent technology assessment service firms, Principled Technologies, just updated their site with a ton of white papers covering performance on just about any product or technology.  You can check out their website at http://www.principledtechnologies.com/default.htm for the complete list, but I wanted to give you the links to their papers that cover virtualization performance on blade servers.

These are the newest white papers, and yes they are mostly Dell.  However, keep an eye on

http://www.principledtechnologies.com/Topics/servers.htm as I’m sure they’ll have HP, IBM and Cisco papers up soon.

    

	Database server consolidation with an Intel Xeon processor E5-2690-based server We tested to see how many dual-socket quad-core legacy servers this new server could consolidate using virtualization.
	Server consolidation and TCO: Dell PowerEdge M620 vs. HP ProLiant BL490 We examine the number of virtual machines each server could support and the total cost of ownership for each server over 3 years.
	Server consolidation and TCO: Dell PowerEdge M620 vs. Dell PowerEdge M710HD We examine the number of virtual machines each server could support and the total cost of ownership for each server over 3 years.
	Server consolidation and TCO: Dell PowerEdge M620 vs. Dell PowerEdge M610 We examine the number of virtual machines each server could support and the total cost of ownership for each server over 3 years

Kevin Houston is the founder and Editor-in-Chief of BladesMadeSimple.com.  He has over 15 years of experience in the x86 server marketplace.  Since 1997 Kevin has worked at several resellers in the Atlanta area, and has a vast array of competitive x86 server knowledge and certifications as well as an in-depth understanding of VMware and Citrix virtualization.    Kevin works for Dell as a Server Specialist covering the Global 500 East market.

HP launched the new  ProLiant BL460c Gen8 blade but you would have never known about it.   With the prelaunch work done it February it seemed like a non event.   The bulk of the goodness on the Gen 8 updates are across the ProLiant  family so I will try to keep my focus on those that pertain to the  BL460c.

The BL460c is focused around the new Intel Xeon E5-2600 Family  processors with up to 8 cores/16 threads each, advanced memory capabilities and new cloud ready management .

CPU
Eight-Core Processors

Intel® Xeon® E5-2680 (2.70GHz/8-core/20MB/130W) Processor
Intel® Xeon® E5-2670 (2.60GHz/8-core/20MB/115W) Processor
Intel® Xeon® E5-2665 (2.40GHz/8-core/20MB/115W) Processor
Intel® Xeon® E5-2660 (2.20GHz/8-core/20MB/95W) Processor
Intel® Xeon® E5-2650 (2.0GHz/8-core/20MB/95W) Processor
Intel® Xeon® E5-2650L (1.80GHz/8-core/20MB/70W) Processor

Six-Core Processors

Intel® Xeon® E5-2667 (2.90GHz/6-core/15MB/130W) Processor
Intel® Xeon® E5-2640 (2.50GHz/6-core/15MB/95W) Processor
Intel® Xeon® E5-2630 (2.30GHz/6-core/15MB/95W) Processor
Intel® Xeon® E5-2630L (2.0GHz/6-core/15MB/60W) Processor
Intel® Xeon® E5-2620 (2.0GHz/6-core/15MB/95W) Processor

Quad-Core Processors

Intel® Xeon® E5-2609 (2.40GHz/4-core/10MB/80W) Processor
Intel® Xeon® E5-2603 (1.80GHz/4-core/10MB/80W) Processor

Dual-Core Processors

Intel® Xeon® E5-2637 (3.0GHz/2-core/5MB/80W) Processor

One interesting thing I saw today when trying to get find out more  information was that you need both processers installed to use both PCIe 3.0 slots. Just something to keep in mind.

Memory

New HP SmartMemory is qualified to run at speeds up to 25 percent faster than industry standards but there is a catch. HP memory from previous generation servers are not qualified to run with Gen 8 Blade.  The HP SmartMemory has to pass testing by HP before meeting qualification. The best part about the new architecture is not having to guess if the amount of memory you put in the server will reduce the overall speed. It use to be if you filled all three DIMMs per channel performance would drop. The ProLiant BL460c Gen8 has four memory controllers per processor socket so that memory speed and capacity is not affected as more memory are added. When using the correct combination of processors and DIMMs, the server supports operating all 16 DIMMs at 1600MHz providing up to 256 GB for maximum memory speed or 1333MHz providing up to 512 GB for maximum memory capacity.

Standard (Pre-configured Models)
64GB (8 x 8GB) DDR3 1600MHz RDIMMs at 1.5V
32GB (4 x 8GB) DDR3 1600MHz RDIMMs at 1.5V
32GB (4 x 8GB) DDR3 1333MHz RDIMMs at 1.35V
16GB (4 x 4GB) DDR3 1333MHz RDIMMs at 1.35V

Maximum (LRDIMM)
512GB (16 x 32GB) up to 1333MHz at 1.35V
Maximum (RDIMM)
256GB (16 x 16GB) up to 1600MHz at 1.5V
256GB (16 x 16GB) up to 1333MHz at 1.35V
Maximum (UDIMM)
128GB (16 x 8GB) up to 1333MHz at 1.35V

The maximum RAM capacity is only based on 16 GB DIMMS but 32 GB will follow soon. There was even a non HP announcement today about 64GB DDR3 LRDIMM’s.

Cloud Management

The new management capabilities for Gen 8 have seem to come along away. In the end all the server vendors are playing with Intel , so this area is probably where the most vendors will differentiate themselves . My favourite new feature is the HP Active Health System.
All configurations change are logged, failed parts are tracked so they don’t get but back in and all the Active Health information can be sent to support automatically so I don’t have to stay on the phone with support.

The cloud based management is based on HP Insight Online. The portal gives you access to monitor your servers, warranty and contracts, service tickets and allows for replacement parts to be proactively ordered. You can give your partners access your portal for remote management and but it looks like automatic (direct-to-expert) support will cost you some additional dollars.

It’s also worth noting that HP changed their drives to better report and show drive failures. The old hard drives will not work in the new
blades.

As more information is released I plan to write about the impact for VDI on my blog at ITbloodpressure.com.

Dwayne is the newest Contributor to BladesMadeSimple.com and is the author of IT Blood Pressure (http://itbloodpressure.com/) where he provides tips on Virtual Desktops and gives advice on best practices in the IT industry with a particular focus in Healthcare.  In his day job, Dwayne is an Infrastructure Specialist in the Healthcare and Energy Sector in Western Canada.

I once was a huge fan of HP’s Virtual Connect Flex-10 10Gb Ethernet Modules but with the new enhancements to VMware vSphere 5, I don’t think I would recommend for virtual environments anymore. The ability to divide the two onboard network cards up to 8 NICS was a great feature and still is, if you have to do physical deployments of servers. I do realize that there is the HP Virtual Connect FlexFabric 10Gb/24-port Module but I live in the land of iSCSI and NFS so that is off the table for me.

With vSphere 5.0, VMware improved on its VMware’s Virtual Distributed Switch (VDS) functionality and overall networking ability, so now it’s time to recoup some of that money on the hardware side. The way I see is most people with a chassis full of blade servers probably already have VMware Enterprise Plus licenses, so they are already entitled to VDS, however what you may not have known is that customers with VMware View Premier licenses are also entitled to use VDS. Some of the newest features found in VMware VDS 5 are:

The last feature is the one that makes me think I don’t need to use HP’s Flex-10 anymore. Network I/O control (NIOC) allows you to assign shares to your network interfaces set priority, limits control congestions all in a dynamic fashion. What I particularly like about NIOC as compared to Flex-10 is the wasted bandwidth with hard limits. In the VDI world, the workload becomes very bursty. One example can be seen when using vMotion. When I’m performing maintenance work in a virtual environment I think it sure would be nice to have more than 2 GB/s a link to move the desktops off – however when you have to move 50+ desktops per blade you have to sit there and wait awhile. Of course, when this is your design, you wait because you wouldn’t want to suffer performance problems during the day by lack of bandwidth on services.

A typical Flex-10 configuration may break down the on board nic (LOM) something like this

To get a similar setup with NIOC it may look something like this.

Total shares from above would be: 5 + 50 + 40 + 20 = 115

In this example FT, iSCSI and Replication don’t have to be counted as they will not be used. The shares only kick if there is contention. The shares are also only applied if the traffic type exists on the link. I think it would best practice to limit vMotion traffic as multiple vMotions kicking off could easily exceed the bandwidth. I think 8000 Mbps would be reasonable limit with this sort of setup.

Management: 5 shares; (5/115) X 10 Gb = 434.78 Mbps

NFS: 50 shares; (50/115) X 10 Gb = 4347.83Mbps

Virtual Machine: 40 shares; (40/115) * 10 Gb = 3478.26Mbps

vMotion: 20 shares; (20/115) X 10 Gb = 1739.13Mbps

I think the benefits plus the cost saving is worth moving ahead with a 10GB design with NIOC. Below are some list prices taken on November 28, 2011. Which one are you going to choose?

Flex-10

HP_6120G-XG

Dwayne is the newest Contributor to BladesMadeSimple.com and is the author of IT Blood Pressure (http://itbloodpressure.com/) where he provides tips on Virtual Desktops and gives advice on best practices in the IT industry with a particular focus in Healthcare.  In his day job, Dwayne is an Infrastructure Specialist in the Healthcare and Energy Sector in Western Canada.

Hot off the presses is the latest IDC worldwide server market revenue report for Q3 2011. The gist of the report is that while some of the numbers are slightly adjusted, really not much has changed in the blade server market.

Revenue growth for the entire server market (all servers, not just blade servers) slowed considerably showing only 4.2% year over year growth bringing in $12.7 billion. Growth in the world of servers continues but this marks the slowest growth rate for any quarter since Q1 2010. IDC believes that overall server sales will continue to decelerate due to weakening economic conditions around the globe. “After nearly two years of steady revenue growth, the server market began to decelerate in Q3 2011 as demand stabilized for many system categories,” said Matt Eastwood, group VP and general manager. Incidentally, IBM and HP are both holding steady, tied for the #1 spot in revenue share, at 29.8%.

When looking at the blade server market specifically, growth was steady for Q3 2011 but not as explosive as Q2 2011. IDC reports “solid growth” in the quarter with a revenue increase of 16.4% year over year (vs 26.9% growth in 2Q11). Shipments increased 2.4% (vs 6.2% reported growth for 2Q11). One thing that hasn’t changed since last quarter is that 89% of all blade revenue is driven by x86 systems. Also, blade server sales representing  20.8% of all x86 server revenue. This shows continued steady growth for the blade server segment but that the pace may be slowing slightly.

#1 market share: HP managed to hold the majority margin moving to 51.0% in Q3 2011 from 51.9% in Q2 2011.

#2 market share: IBM continues to see its margin chipped away slightly down to 18.5% in Q3 2011 from 19.1% in Q2 2011.

#3 market share: Cisco’s disruptive market penetration seems to have slowed at 10.7% overall compared to a solid 10% in Q2 2011.

#4 market share: Even Dell dropped slightly to 7.2% revenue share from 8.2% last quarter.

In looking at the totals, the top four vendors represented 87.4% of the revenue share in the blade servers market which is actually down 2% from last quarter. Cisco grew revenue share by less than 1% which means that some of the displacement of the remaining top vendors is not accounted for. Does this mean there may be some new players in the “others” category that we should be watching? Without a detailed breakdown it’s hard to tell but I’ll definitely be looking forward to comparing the numbers next quarter to see if the trend continues. It could, after all, just be a factor of the margin of error in the statistics.

According to Jed Scaramella, research manager, Enterprise Servers at IDC, “Blade systems represented the fastest growing segment in the server industry and now account for 16.0% of total server revenue – a historic high.”

Probably the most interesting aspect of the report is the introduction of hyper-scale servers. “Hyper-scale servers are designed for large scale datacenters with streamlined system designs that focus on performance, energy efficiency, and density.” This sounds like the mantra for blade servers with the main difference being the lack of management and high availability capabilities at the hardware level. Basically these represent the miles of simple, rack mount commodity servers used by the likes of Google and Facebook. This is a $428 million dollar server segment and growing.

For the full IDC report covering the Q2 2011 Worldwide Server Market, please visit IDC’s website at http://www.idc.com.

With all that is made of the competition between blade server manufacturers and the growth of the blade server market in general, is there room for another type of condensed computing in the data center? Have we been going about things all wrong with regard to architecture design?

Nutanix thinks so.

Nutanix is a start-up company geared towards delivering a simplified virtualization infrastructure with a strong focus towards eliminating the need for a SAN. Their clustered solution brings storage and compute together which theoretically reduces expense, reduces complexity, and improves performance. On its own it doesn’t really seem that innovative but the secret sauce is how they make the cluster scale and tier/span data across all nodes without sacrificing performance. Each node has the usual compute resources plus a mix of local SSD and SATA hard disks. There are 4 nodes per 2u enclosure called a “block”. Add more blocks and you have a Nutanix cluster. The software stack scales and balances everything between the nodes and blocks. The technology originated from the architecture that companies like Google and Facebook employ in their data centers. Assuming that can be taken at face value, the scalability potential is phenomenal.

So what’s the big deal?

Well my thinking is that if you can eliminate the need for a SAN (for virtualization) then you can definitely eliminate the need for an enclosure of blade servers. No interconnects. No Enclosure. Simplified network architecture. No SAN. What’s not to love?

Eliminate the SAN. Really?

The 37 million dollar question that comes to mind is: Can I really eliminate my SAN? For me the answer is a quick and resounding, “No I cannot!” My SAN does a heck of lot more than deliver servers through a hypervisor. There are still application, database and file servers depending on a direct connection to that centralized beast of disk technology. Many of those servers may be virtualized but some of their data volumes are simply best served as LUNS and not ridiculously large VMDK files. Granted, the topic of VM data placement is debatable but it becomes moot if I have any physical servers around needing centralized storage. In my environment, and I feel confident I’m not alone in this, the SAN can’t really go away.

Specific to virtualized workloads, Nutanix claims that there is plenty of capacity for just about any scenario. Personally, I still doubt the clustered Nutanix nodes have the local capacity to absorb both the VMs and their data with redundancy. A single block (made up of 4 nodes) has some solid state goodness and 20TB worth of storage (5TB per node). As the cluster scales by adding blocks of nodes the increased pool of storage is available to all VMs. For some environments that may be more than enough capacity to contain everything related every virtual server. For other environments this may not even be close to the needed capacity. I also wonder what happens when I drink the Nutanix kool-aid and then later hit the capacity wall? I’d feel pretty silly running out of space and having to buy a full compute block just to increase storage capacity.

Regardless, it does seem possible that Nutanix could displace the SAN completely if your environment is 100% virtualized and capacity requirements aren’t too drastic. Realistically though, since most businesses are not 100% virtualized, this technology simply reduces the need for capacity expansion and the worry of performance tuning the SAN for VM workloads. That definitely has some value but how much depends on how virtualized you are and how much data your VMs are connected to (among other things).

Ok. So how about compute?

Today we have stacks of blade enclosures full of ESX hosts driving maximum efficiency of network and power. Or are they? Let’s say you have a HP c7000 full of half height BL460 servers. For every 10u of rack space you get 16 dual socket ESX hosts. With the Nutanix solution there are 4 hosts per every 2u block so that same 10u can hold 20 similarly configured physical hosts. Within that same 10u there is also a collective 6TB of Fusion IO, 6TB of SATA SSD, and 100TB of SATA disk. The density win clearly goes to the Nutanix solution.

From a networking perspective things are a bit less clear. Within the HP enclosure, using Flex-10 virtual connect, there would be two or more 10Gb uplinks per enclosure (16 physical hosts). For each pair of uplinks one would be active with the other ready for fail-over. You could get away with a single pair of 10Gb uplinks but that would be silly for a full enclosure of hosts. I’d argue that more than four uplinks would likely be needed but that’s a different conversation. The main point is that there is some flexibility and design decisions as to how the network is configured.

By contrast the Nutanix has four 10GB uplinks per 2u block (one for each node within the block). Comparing the same 10u rack space you end up with 20 10Gb uplinks. Nearest as I can tell there is no aggregation of the 10Gb interfaces. From a pure performance perspective you can’t argue with the simplicity and potential of a dedicated 10Gb connection per host. On the other hand that is a huge number of 10Gb ports and certainly increases the cost of the networking infrastructure. The single interface per node also doesn’t give any option for fail-over. At this point I’m not entirely clear on what all of the networking capabilities are because Nutanix literature focuses purely on all of the SAN oriented benefits and I haven’t yet connected with them to follow up. I’d like to learn a bit more about their networking architecture before I declare an advantage to one or the other. I’ll report back what I find out.

Bottom Line

Nutanix has a pretty good start on a new way of delivering virtualization. The idea of consolidating compute and storage isn’t entirely new (Pivot3, for example, has had VMs running on their storage arrays for a couple of years now) but Nutanix has absolutely evolved the concept. If they can deliver on all of their promises for less money than “the big guys” they stand a reasonable chance at being disruptive to both storage and blade center markets in the niche of virtualization.

One area where I can see this strategy really taking off is with VDI deployments. Who wouldn’t want a dedicated stack of gear running at high performance with plenty of capacity to spare? The total value of this technology,however, depends on a number of unanswered questions both technical and financial. Assuming it is as cost competitive as it is claimed to be, certainly green field environments should take a serious look and determine if a dedicated purpose built virtualization cluster has more value than a more complex but more versatile traditional storage and compute approach. Those that have existing storage and compute stacks will need to do a much more comprehensive evaluation because there’s a lot more to the value of “traditional” data center technology than consolidated server workloads.

Nutanix is cool, but I’m not replacing all of my blade servers just yet.

To find out more information about Nutanix click here.

To find out more about HP blade servers click here.

Chris is a contributor and author for BladesMadeSimple.com and in his day job is a Systems Architect for Clackamas County in Oregon and an HP customer.  The opinions expressed in his writings do not necessarily reflect those of BladesMadeSimple or Chris’ employer.  Follow Chris on Twitter @sysgeekguy.

People are always asking me: “how many blades I have to put in a chassis to make it worth?” My answer is always: “It depends!” . There is no magical number.

First of all, you have to buy* a chassis and since Blade servers are cheaper than rack servers (considering the same configuration), you can save on TAC (total acquisition cost) depending on how many blades you put into each chassis.

Also, even if the customer wants just a couple blades, depending on how many I/O modules you will put into a chassis, it could be cheaper than a complex solution involving two or three vendors and less discount.

This comparison is just about TAC, I will not get into all the benefits of a Blade environment.

Rack servers will have advantage for a customer that has the I/O  infrastructure and wants to buy just a few servers and not be tied to a vendor specific technology.

*Both HP (post here) and IBM (post here) have promotions that give free Enclosures/Chassis on your first buy.

My experience says that is better if a customer buy only the minimum necessary to apply for the promotion and request a discount on the rest of the configuration.

About the Author
Thales Osterne is a contributor and author for BladesMadeSimple.com.  He has over six years in the IT field with four years of experience in IBM BladeCenter and System X.  When he is not blogging, Thales works as a product manager for IBM System X & Bladecenter at Lanlink Informatica, a major business partner in Brazil.  He is fluent in Portuguese and English.

Not long ago it was reported that IBM was having a promotion to receive a free blade enclosure if you are a first time customer and the chassis in question is eligible. Details can be found in the following links:

Link to blog post about IBM offer
Link to actual IBM offer

More recently we posted info that HP was offering something very similar. The main difference was that HP’s promotion requires that you purchase some hardware to get the free enclosure.

Link to HP promotion

Since I’m a HP blades customer and I’ve been planning on adding some additional blade servers soon I thought I’d better follow up and get details on the whole deal. What I found out was a bit disappointing.

Basically the blades you have to buy in order to get a “free” enclosure are pre-configured and you cannot customize them when you order. Specifically, they are DL460 G7’s and only have a single processor (Intel® Xeon® X5675 (3.06GHz/6-core/12MB/95W, DDR3-1333) with 12GB of RAM and no HDD. If you don’t like that configuration you’ll need to buy the additional or replacement parts separately and install them yourself. That may not be a deal breaker but still a bit annoying that the configuration is so rigid.

The enclosure, on the other hand, was configured better than I expected. There are no  interconnect modules (naturally) but the c7000 chassis comes with 6 power supplies, 10 Fans, 16 ICE Licenses and is setup for single-phase power (6 power cords out the back). The chassis configuration, such as it is, is also not configurable so if you use three phase power or direct DC current then you’ll be changing out the power connector or adjusting your circuits to match.

It’s a shame that the configurations are not adjustable and I’m unsure whether the cost savings will be worth the hassle of mix-matching parts to get the configuration I actually want. I don’t have any pricing yet but once I find out the general difference between buying the blades and chassis separately through my own channels versus the special promotion I’ll let you know if it’s a deal worth looking into.

Till next time,
Chris Fricke

Chris is a contributor and author for BladesMadeSimple.com and in his day job is a Systems Architect for Clackamas County in Oregon and an HP customer.  The opinions expressed in his writings do not ncessarily reflect those of BladesMadeSimple or Chris’ employer.  Follow Chris on Twitter @sysgeekguy.

In case you haven’t heard, “cloud” discussions are here to stay and everyone has their own recommendations for you to start building your cloud environment, but which is best for you?  Intel has created the “Intel® Cloud Builders” program aimed at making it easier for you to build, enhance and operate cloud infrastructure.

The Intel® Cloud Builders site is a collection of Cloud Design architectures resources from all of the major vendors like Cisco, EMC, Dell, HP and IBM.  Once you understand your cloud requirements then you can filter through the architectures based on one or more of the following:

1) Usage Models

• Infrastructure as a Service (IaaS)
• Policy-Based Power Management
• Scale-out Storage
• Trusted Compute Pools
• Unified Networking
• Client-Aware (self-service)

2) Cloud Computing Categories

• Build and Simply Your (existing) Cloud
• Improve Efficiency
• Enhance Security

3) Preferred Ecosystem Partner (allows you to pick your preferred hardware or software vendor)

While the site provides useful information, it is still limited.  For example a search in “all Dell” within the Content Library returns 3 documents; “all Cisco” returns 1 document and “all HP” returns 2 documents.  I see really good potential for this site, but Intel and their Ecosystem Partners are going to need to provide more documentation.  I would also like to see Intel introduce a filtering category that allows you to choose between rack servers and blade servers. 

Overall, I would give the site a 3 out of 5, so if you are looking for a resource to help you with your cloud design, check out Intel® Cloud Builders at http://www.intelcloudbuilders.com/

I was recently on one of those calls from HP “checking in”. If you are an HP customer you may know these calls start off with something like “We know you spent money on HP last quarter so how many people from this company will have to pester you until you buy more stuff from us?” One of the things the HP rep mentioned towards the end the process, however, is that until October HP is now offering a free blade enclosure with the purchase of four HP BL460′s.  I have to admit, I haven’t talked to anyone about the gritty details yet but this sounds awfully familiar to the promotion that IBM is having that was reported here earlier this month. I did some research online and found a few links with various versions (or at least descriptions) of this offer including one for UK customers that is about to expire:

Link 1
Link 2

Incidentally, I did some poking around directly on the HP site and found no advertisements or even mention of the “free chassi” promotional offer.   It seems to be a weak promotion if there’s nothing to indicate that it even exists. Regardless, does this mean that Dell is next? Does this mean that Cisco UCS is more disruptive than the HP, IBM or Dell are letting on or is it simply a case of corporate “me too”? Seems like a silly bandwagon to jump on given that most of the cost is not in the chassis, but in the I/O modules. Give away a free enclosure with a pair of Flex-10 modules installed and then you’ll have my undivided attention, Mr. HP. Kudos.

Chris is a contributor and author for BladesMadeSimple.com and in his day job is a Systems Architect for Clackamas County in Oregon and an HP customer.  The opinions expressed in his writings do not ncessarily reflect those of BladesMadeSimple or Chris’ employer.  Follow Chris on Twitter @sysgeekguy.

Many people get confused as to why so many I/O modules are needed within a given blade chassis.  The basic concept is simple (in most cases) – for each port you need on a given blade server, you need to have a corresponding I/O module.  For example, if you need 4 NICs, you’re going to need 4 Ethernet modules (in most cases.)  In today’s post, I thought I would keep it simple and publish the I/O diagrams of Cisco, Dell, HP and IBM chassis.  Of course, I am human and “have been known to make mistakes – from time to time” so please feel free to correct me on any errors you see.  Enjoy.

(Updated 8/3/2011 – fixed Dell M1000e Full Height I/O Diagram)