VMware Support updated

February 18th, 2010 by dawoo No comments »

VMware have updated their support model:

Gold Support will be renamed Basic Support
Basic Support is for test, development and non-production environments, where basic response time targets and Business Day support is acceptable.

Platinum Support will be renamed Production Support
Production Support is for production environments, where faster response time targets are needed, and 24 x 7 assistance for Severity 1 issues is required.
 
Business Critical Support name remains unchanged
Business Critical Support is an add-on to Production Support and is for those customers who have deployed VMware technology in their critical production environments where support from a designated technical support team is required.

I can’t imagine you’d really need a Business Critical contract given the stability of their products!

http://www.vmware.com/support/services/overview.html

VMware Sizing Tools

February 12th, 2010 by dawoo No comments »

During my current design I opened by usual bookmarks to start my capacity planning exercise for vCenter, Update Manager etc… Anyway, I thought I’d share the URLs and if anyone has any they’d like to add I’ll keep this entry updated.

VMware Update Manager

 http://www.vmware.com/support/vsphere4/doc/vsp_vum_40_sizing_estimator.xls

VMware vCenter

 http://www.vmware.com/support/vsphere4/doc/vsp_4x_db_calculator.xls

VMware Site Recovery Manager

 http://www.vmware.com/files/pdf/Site_Recovery_Manager_4.0_Database_Sizing_Calculator_SQL.xls

VMware 3.5 Update 5 supports Windows 2008R2 & Windows 7

January 15th, 2010 by dawoo No comments »

Following on from my previous post…

If you’re running VMware ESX 3.5 you’ll need Update 5 if you’re planning to run Windows 2008R2 and Windows 7 guest operating systems.

 Release notes: 

 

http://www.vmware.com/support/vi3/doc/vi3_esx35u5_rel_notes.html

vSphere 4 Update 1 supports Windows 2008 R2

December 2nd, 2009 by dawoo 1 comment »

This is not breaking news but moreover a ‘note to self’ for anyone with the same thought process as me.

VMware training and common practice always says you should check the VMware HCL for any compatibility checks. Of course, what happens when the HCL isn’t up to date? It happens.

MS Windows 2008 R2 is listed as Experimental in the HCL for vSphere 4 Update 1 but actually the release notes state it’s officially supported.

So the ‘note to self’ for today…

“Also check the Release Notes”

Service delivery with resource pools

November 27th, 2009 by dawoo No comments »

Where do you start when designing Resource Pools in an empty VI?

a) Create pools based on Development, Test & Live then divide total resource by 3?
b) Divide total resource by a rough split of your planned P2V phasing?
c) Create pools for logical grouping of VMs?
d) Group by vCPU or RAM allocation?

etc… Anyway, I think I’ve painted the picture. So what’s the answer?

Simple, understand the business needs.

“But that’s not a technical answer!”

A Virtualised Infrastructure is really no different to any other service provision.  If you’ve just moved home you wouldn’t ring up your local water authority and request ‘x’ amount of water must be provided to your home. You’d say “I’d like water please and let me know how much I owe you for a residential supply.”

Why not apply this methodology to service provision?

Example, you need a single SMTP relay running under IIS 6, anticipated 10k mail a day.  POC this with a test tool, review utilisation demands.  Snapshot results, add a bit more based upon forecasted peeks, done.  The project now requests what it needs to deliver the requirements.

Resource pool decisions don’t necessarily have to be final.  Shape the delivery of services around business demand.  Force new projects to fully comprehend their requirements and impact.

We’re in a strong position using a Virtualised Infrastructure:
- Resource can be ramped up, down and constrained. Whereas fast executed projects using physical service provision can be disastrous if poorly executed.

Resource allocation with a sliding scale – it’s the way forward.

Best Practice vs Fit for Purpose

November 13th, 2009 by dawoo No comments »

Reading this article  (Protocol Passionistas) on ViewYonder it prompted me to write up this article which, I believe, are of the same mindset.

Hearing the phrase ‘Best Practice’ without including discussion around shaping  the solution to be ‘fit for purpose’ induces concern within me.
There is a myriad of white papers and best practices on the internet  ready for download from vendors & suppliers, all of which contain  invaluable information on how to configure their products.  But it  requires an open minded approach to review their recommendations and  consider what is actually relevant and suitable for the task in hand.
In a site visit earlier this year my friend, and business colleague,  attended to investigate why an application would no longer communicate to  key components of a multi-tiered e-commerce application.  It  transpired that new network hardware had been introduced and followed  a best practice document for security hardening.  This was all well  and good but no investigation has been completed as to what this  actually meant to the existing infrastructure.  The end result?  The application ceased working causing loss of service to business and  their customers.  Not a good way to make a name for yourself and using the excuse “I followed best practise” doesn’t carry much clout.
It’s examples like this where solutions are provided using Best Practice  documents in place of installation & configuration instructions  whereas they should be used in conjunction with installation &  configuration instructions.

Best Practice is not the answer.  Common sense, practical approach and  understanding what is fit for purpose with risk mitigation, is the  answer.

Q: Why shouldn’t you virtualise?

November 12th, 2009 by dawoo No comments »

I posed myself this question on the train recently, “Why shouldn’t you virtualise?”, a somewhat controversial question coming from me as I’m completely on the flip side of that question.  However, a few thoughts provoked this query.

Why not?

  • If your physical world is a headache (other than hardware failures) then take a step back.
  • Virtual Machines don’t manage themselves, other than host resource control.
  • Operating System management doesn’t miraculously happen.
  • In the wrong hands (literally) the mouse right-click can be the enemy for uncontrolled rapid cloning and deployments of virtual machines adding to the sprawl.

Virtualisation core elements:

1. Knowledge
Desktop virtualisation is widely used but data centre editions are different.

2. Comprehension:
It’s not about why a Type 1 hypervisor is core for the enterprise data centre but how to genuinely configure it to meet service level demands.

3.  Appreciation:
The core components of a VI align to physical infrastructure
Physical infrastructure is core to provision
VM deployment should be controlled and audited – the same as physical

 Immaturity of a VI can lead to:

  • The managment & administration delegation simply being ‘All Admins’.
  • Virtual machines mirroring the physical world – prior to the consolidation exercise of course.
  • Ad-hoc virtual machine provision without thought for impact.

Answer?  Service provision maturity is not solved by virtualisation technology, it’s a mind set.

Wordpress iPhone App

November 10th, 2009 by dawoo No comments »

The Wordpress iPhone app is fast becoming a popular download. Rather than purely Tweet about it what better way to test it out than blog about it using the App!

So there we go.

HP Proliant DL380G6 – 18 DIMM Population solution

November 9th, 2009 by dawoo No comments »

While pulling together a build document for an ESX upgrade project I came upon an interesting challenge regarding DIMM placement and also established a few good resources worthy of future reference.

Anyway, back to my build.  Equipment previously spec’d by a predecessor arrived on my desk. 14 x HP Proliant DL380G6 Dual CPU with 60gb RAM, I won’t elaborate on other hardware aspects as that’s not the point of this item.  And yes, I have a big desk.

The 60gb comprised of 6 x 2gb and 12 x 4gb DDR3 Dual Rank DIMMs.

The 3rd party supplier built server arrived with the RAM positioned:

3rd Party shipped

incorrect-placement

There is no doubt as to whether the server would work but I was keen to understand the optimum performance achievable by following manufacturer recommendations.

The DL380G6 uses the Xeon 5500 architecture and has 3 memory channels per CPU, channel 1 being ‘closest’ to the processor.

The HP documentation suggests the population order should follow the alphabetical notation and not the slot number.  To further confuse matters the order of population starts from the white DIMM slot.

For those of you not having seen inside the server the white DIMM slots are numbered 3, 6 and 9.

This equates to :

Channel 1 using slots 3 + 6 + 9

Channel 2 using slots 2 + 5 + 8

Channel 3 using slots 1 + 4 + 7

HP DL380G6 System Board

SYSBOARD

Which, in memory channel terms looks like this…

RAM, CPU and Memory Channel Explanation

RAM

HP state that memory should be balanced across memory channels and CPUs, the original build only met one of the criteria for an optimal build.

Here’s the finished placement following HP guidelines.

HP correct placement

correct-placement

Resources reviewed to assist with my conclusions

This image was taken from page 17 of the HP Proliant DL380 G6 Server User Guide, the document is located here:

http://bizsupport1.austin.hp.com/bc/docs/support/SupportManual/c01705681/c01705681.pdf

Populating the 18 slots has been taken from HP the recommendation in Table 6 Page 10 of this document:

ftp://ftp.hp.com/pub/c-products/servers/options/Memory-Config-Recommendations-for-Intel-Xeon-5500-Series-Servers-Rev1.pdf

HP DDR3 Memory configuration tool:

http://h18004.www1.hp.com/products/servers/options/tool/hp_memtool.html

Technical: How to Virtualise your organisation

August 25th, 2009 by dawoo No comments »

START : Insert VMware ESX DVD, click next few times then reboot.
FINISH : That’s it, right?

Yes, if you :
… are trying to keep with industry trends.
… consolidate server sprawl to meet your “Green IT Initiative” objective.
… consider virtualising just the ‘quick win’ candidates.
… continue to manage & maintain using your existing infrastructure procedures.

No, if you consider there’s opportunity for :
… mapping current Virtualisation Infrastructure solution potential to your business.
… shaping delivery of Infrastructure services to the business.
… promoting an ‘Agile & Can-do’ image of IT services to the business.
… taking time to review your infrastructure provision.
… dropping the ’server per project’ mentality trap many have fallen into.
… shape new TCO & ROI models projected over time mapping to business capital & operational expenditure.

What about the bigger picture?
The picture has the potential to evolve exponentially given an open approach. Designing for today’s business demands is tantamount to setting yourself to fail in a month, 6 month or years time. Thinking outside of the initial remit, involving business units regardless of technical ability promotes the accepted idea of change, increases confidence, encourages learning and assists in reducing sponsorship barriers. After all, spend is required.

In summary
Implementing a Virtualisation solution is not about installing the latest infrastructure technology or re-inventing the infrastructure provision wheel. It’s about creating an adaptive infrastructure using today’s technologies, applying proven frameworks and methodologies with the goal to deliver over and above business expectation.

Of course, this is just my opinion.