8 GBIT/SEC - HBA and Switches

Brocade announces 8 GBIT/SEC - HBA and Switches. (PDF)

It's all about the ratio.

VKernel is an advocate of running your hardware at high levels of capacity. I know that we would see record-breaking ratios of virtual machines to server hardware.

• A major worldwide financial services organization achieved a 12:1 consolidation ratio and increased its central processing unit utilization by 30 percent.
  
  
• An Indian petroleum refining and distribution company achieved a 17:1 consolidation ratio and expects to increase that to 30:1 with additional CPUs and RAM.
  
  
• One of Italy's largest banks improved its server utilization rates by 100 percent.
  
  
• A leading US faucet manufacturer saved $250,000 in hardware costs by reallocating existing units instead of purchasing new, achieving a 10:1 consolidation ratio.
  
  
• A South American energy company consolidated its servers by a 20:1 ratio.
  
  
• A federation of trade unions in Singapore consolidated its servers by 46:1, achieving a 26 percent savings.
  

Smallest is 10:1 and largest is 46:1.

It's all about the ratio.

Miles Per Gallon for Data Centers

At the heart of VKernel is the notion that you can't manage what you don't measure.

I wrote earlier in the week about Will Forest (McKinsey) who is advocating a CAFE-like standard for Data Centers and the MPG analogy works.

If you are running your VMWare at lower capacity levels - its driving a car with a low MPG, increase the capacity and you increase your MPG.

Forest posits CADE - Corporate Average Data Efficiency or MPG for servers.

Take a look at this graphic for CAFE - Fuel Efficiencies

In Cars (and or light trucks) - its MPG vs. Footprint (Sq. Ft).

As they increase in square footage, their MPG decreases - smaller vehicles would have better fuel economy/efficiency but not be able to transport goods/people - so their ROI might be tricky if they had to haul sheets of plywood.

In Servers (not light trucks) - its Servers vs. Virtual Machine Density.

IT shops running the early versions of server hardware are going to consolidate to increase ROI, groundbreaking virtual machine density numbers.

“It clearly makes more sense to become more efficient than to build another $100 million data center,” said Kenneth Brill, executive director of the Uptime Institute.

Be more efficient, drive up capacity, save the planet.

If data centers were hotels...

Steve Lohr over at the New York Times has written a piece about the fact that servers are massively underutilized and if they were hotels they would be bankrupt.

He wrote that "computer servers are used at only 6 percent of their capacity on average, while data center facilities as a whole are used at 56 percent of peak performance."

This is one of the biggest selling points about virtualization - drive up utilization of the server hardware.

The piece is based on a McKinsey & Co. study by Will Forest who says "world’s data centers are projected to surpass the airline industry as a greenhouse gas polluter by 2020."

Okay great - so its not just a capacity issue, now its an "save the planet" issue.

Another of the big selling points of virtualization is the more efficient use of power, cooling, etc.

(Ed. note: The word environment can get confusing when in a server virtualization discussion - anything that has to do with environment in this blog post is related to server power and the heat they give off, cooling of said heat and the powering the cooling. (floorspace, voltage, amps, etc.???).

It's these resources that are being consumed inefficiently when the servers are running at low capacity and directly proportional - make these boxes run at higher capacities, drives down the environmental requirements and reduces the per VM costs.)

He states (via Lohr) that the negative environmental impact will come from the electricity consumed by the growing data centers and the "the carbon dioxide emissions attributable to the electricity consumed by fast-expanding data centers will rise fourfold, the study estimates".

McKinsey states (via Forest) that to help fix this IT Departments will need to "bring some of the mainframe-style management disciplines to modern data centers".

Amen. Amen. Amen.

(Ed. Note: That's our first three amen blog post).

Virtualized servers as mainframes - that's genius talking right there.

And that means big iron management, capacity management, better systems management than what existed in the past - its mainframe disciplines brought to virtualized servers.

VMs can cost more than physical servers -- really!!

I am amazed how many virtual environments I have now seen that are severely under utilizing the new hardware and are afraid to increase VM density. They buy expensive server hardware, loaded it with 16Gigs or more for $30K to $50K and are running just a handful of VMs on it. This is analogous to driving a perfectly good Ferrari without ever getting out of first gear!! Say your are running 8 Vms on a $50K hardware. Add the cost of SANs, etc and you can quickly see how the cost of each VM can actually be higher than the physical server it replaced. This of course begs the question why do people underutilize the hardware?

As far as I can tell there are several reasons. Some are just being utilization "ignorant" about their environment, but the majority is simply afraid to "push the metal" and increase utilization because of concerns about running into ESX performance problems or worth yet -- downtime. Since finding capacity bottlenecks using Virtual Center is not trivial and time consuming, and predicting future capacity bottlenecks requires fairly advanced mathematical analysis of all core 4 resource types , disk I/O etc, most Vmware Admins lack the time and experience to do this exercise. So they keep the Ferrari in first gear, keep driving blindfolded, and hope that vm sprawl does not catch up with them. With availability of tools like the Vkernel Capacity Bottleneck Analyzer
VMware admins will gain visibility into current and future capacity problems and steer clear of performance issues. It heps driving with lights on!! Tell us what you think www.vkernel.com

Gartner Identifies Six Best Virtualization Practices.

David Marshall's has an item from Gartner on six things every company should consider before they virtualize.

1. Start Small, Think Big
2. Require a Rapid ROI
3. Virtualize the Right Applications
4. Define Your Storage Strategy
5. Understand Software Issues
6. Combine Virtual Machines Effectively

I like that John Enck called the market "immature" - he said its still evolving and that pricing is going to change due to competition (I have to think this is Xen and Hyper-V coming to market).

When is it going to be mature - who knows - they say the Virtualization space will be "the most-important trend for servers through 2012" and that they have "have had thousands of client interactions on x86 server virtualization since 2001".

They also said their customers are asking "the most-common questions revolve around best practices for starting a server virtualization project. "

Thomas Bittman comments on the fluid, liquid pool idea of resources when he writes "enterprises will effectively leverage virtualization now and will re-architect their servers to become a more efficient, fluid pool of dynamic capacity"

Cloud, Pool, Grid, etc - all of it is going to require management to be effective.

I want my Availability

I wanted to do a knock off the Dire Strait's tune - Money for Nothing and then change the "I want my MTV" into "I want my Availability" - VKernel will then hire a Dire Straits cover band to perform it somewhere.

Availability vs. Capacity - is one better than the other?

Are there types of Availability?

Virtual John writes about High Availability (HA) vs. Continuous Availability (CA)

"In plain English this means, if one of your hosts in a cluster of VMWare Servers goes away the VMs will reboot elsewhere. Reboot = downtime, so is this high availability? Or just higher availability than no fault tolerance?"

It could be a important misnomer - the VM's with HA will be expected to not have any interruption of business service (enter VMotion) and voila - "I want my Availability".