Wednesday 30 March 2011

Capacity management often benefits from an external perspective

In today's ever changing and dynamic environments the need to respond quickly and accurately to business or customer requirements becomes ever more critical. This combined with complex and heterogeneous technologies and potential resource shortages means that the use of consulting services can save money and provide real value. 

I'm running a webinar which demonstrates how using our range of services has delivered real-term benefits to a variety of clients, of varying sizes and markets. 

Using customer case studies I'll provide an overview of the problems, solutions, and benefits associated with the following areas:
  • Website Health Check
  • Consolidation Assessment
  • ITIL GAP Analysis
  • Merger and acquisition assessment
Join me and find out how we can help you ensure that you are getting the best performance and return on investment from your infrastructure.



Rob Ford
Principal Consultant

Friday 18 March 2011

Will ‘Cloud Bursts’ be as expensive as ‘Train Bursts’?

Hmm, I wonder?  I had to catch a train up to London the other week on business.  A chance came up for a meeting I had been trying to get for some time.  I didn’t want to miss the opportunity, the person I wanted to see was passing through London and I had the chance to grab a couple of hours of their time face to face.  Old fashioned guy that I am, sometimes web meetings and video conferencing just won’t do – you need to meet in person.
Next day I arrived at my local train station, bought a return ticket to London and five minutes later boarded the train.   The cost of the basic ticket I bought was £99.50.
This hurried trip reminded me that I had  to visit London again the following month and that I hadn’t yet sorted my ticket out.  I called the office from the train and got a colleague to go on the Internet and pre-book my return ticket for a month ahead.  They called me back to say it was booked at a cost of £30.  Same trains, same day of the week, no holidays or anything else to confuse matters.  Two identical ordinary working days.  A price difference of over 330%.  Clearly it pays to plan ahead when using the train.
As I sipped perhaps the worst latte ever made, and nibbled a pastry with the consistency of hard rock, I glanced through the windows and watched the clouds drifting by.  Of course, you can’t avoid clouds these days if you’re in IT.  Public, private, hybrid or whatever, some form of cloud implementation is on the horizon for all commercial IT users.  One of the great benefits we are promised is ‘elasticity’, the ability to rapidly add additional resources in the event of a sudden requirement.  What price will we pay for these ‘cloud bursts’ or ‘overdrafts’ as they are known.  My elasticity in responding to a sudden meeting requirement had cost me 330%+ more than if I had planned the meeting in advance.  Will we be paying the same premium for IT resources bought at short notice?
Of course we will.  Emergency facilities like your bank overdraft cost more than planned requests for credit.  Air side shops, excluding the delights of duty free, charge more than high street shops, because there is nowhere else you can buy something before you get on the plane.  Adding in resource from your Cloud provider at short notice will cost more than planned, contracted resource.
To avoid costs running away in what could be a panic or emergency situation for your business, the better you can predict and manage what resources you need and when will therefore be critical.  Capacity management, capacity planning and strategic IT service management disciplines will become ever more important if you are to do this successfully and avoid unnecessary spend.
How high will the premium be?  Time will tell.  If you’ve got an experience of the cost differential between planned and cloud burst costs feel free to share them. 

Andrew Smith
Chief Sales & Marketing Officer

Wednesday 16 March 2011

Capacity Managing Data storage

Implementing capacity management processes for a complex SAN storage environment can be a daunting task. For SAN administrators it typically takes a back seat to more tactical operations like allocating storage, hardware updates, and firefighting.
Determining who is responsible for such a capacity management process can also be a difficult question to answer as there are typically overlapping areas of responsibility and internal politics that complicate the issue. Those who normally do traditional capacity reporting and analysis may not have sufficient knowledge of the SAN environment to effectively report on it.
Once responsibilities are sorted out, you need to clearly define and document your requirements and objectives. A tool, such as athene, should be selected that provides automated trending and exception reporting and a centralized reporting structure. Feeding back the information from this reporting is critical in gaining value from the effort and should be integral to the process.
Data can be looked at from multiple perspectives such as the host, fabric/network, storage controller, and other devices but collecting data from all these sources for centralized reporting can be a major challenge. Some vendors may claim to do it all, but in the end there will likely be many challenges regardless.
Two key area of focus should be response at the host and storage capacity utilization.
Classic I/O intense applications are data warehouses, BI operations, large databases, and busy mail servers.
I’ll be discussing ways to assist the storage administrator in the complex task of managing SAN attached storage at my webinar ‘Capacity Management for SAN Attached Storage’ March 24 – register now and join me.

Principal Consultant
Dale Feiste

Monday 14 March 2011

Cloud - measure what you can, control what you can and don’t worry about the rest.

One significant change with the move to Cloud based systems is that capacity management becomes a much more strategic activity than in the past. 

Analyst groups such as Gartner are promoting this evolution and it is further supported by initiatives such as the ITIL® v3 refresh.  Rather than a purely resource level task, it now needs to be an integral part of how the business chooses what is the best solution, for example between private and public Cloud.
The need to use capacity management techniques to plan requirements in advance will be ever more important.  Buying something in advance is typically cheaper than buying at the last minute. Emergency buying of Cloud resource (cloud bursts) might be easy to do but it is likely to be prohibitively expensive over time.

With service potentially coming from a variety of internal and external sources, guaranteeing service quality becomes both more difficult and more necessary.  From a capacity perspective you won’t be able to measure everything you want in the Cloud so measure what you can, control what you can and don’t worry about the rest. 

Our tools and processes support this open approach and as such are an essential.

We can show you how to increase productivity, reduce costs and what you can measure and control in the Cloud through our interactive online presentations, so give us a call..



Andrew Smith

Chief Sales and Marketing Officer

Friday 11 March 2011

Adding Value to your Trend with Confidence Intervals - continued

To continue where I left off on Monday, let’s see what happens as time goes forward and more measurements become available.



Here’s what happened after the second month of measurements.  Clearly the upward trend is continuing – and look how much closer together the confidence limits are.  While you would not yet bet the farm on what will happen on 1st December, you can now see that the extremes of 35% (on the low side) and 90% (on the high side) are both extremely unlikely – indeed, you can be 95% confident that your measurements show a genuine trend between 50% and 70%, much narrower than previously.

However, through all this, please don’t forget Lesson 1 of my previous set of blogs – Most trends don’t extrapolate for ever – try to work out what the limiting factors may be.  In this example, the limit is probably a straightforward factor such as the number of new users still to be migrated to this application. When all the new users are happily set up, the trend will almost certainly stop rising – or at the very least, it will change significantly.  An external factor like this shows that confidence limits on a trend are only useful if the trend remains more or less constant.

In this case, there’s every expectation that the trend will remain more or less constant.  Here's the trend, with the same confidence limits, predicting utilization between 50 and 60%, on three months worth of data collection.
This looks like a trend we can rely on and certainly gives us more confidence than old fashioned guesswork.



In real life you will probably want to track the peak hour as well as the average hour.  That’s fine – you will end up with a range of confidence for the predicted utilization during the peak hour, and almost certainly a slightly different range of confidence for the predicted utilization during an average hour.


In future blogs I’ll look at some more complicated trending issues, including cases where trending just won’t work, and where you will have to use a different prediction technique such as modeling.


Andy Mardo
Product Manager

Wednesday 9 March 2011

Adding Value to your Trend with Confidence Intervals

Suppose you choose to capture performance measurements from your servers once every 5 minutes, which is a reasonable compromise between losing some detail and drowning in data.  A week or a month later, you can’t go back and ask for more data points during the same time period – the opportunity has come and gone.  Any practical data collection exercise involves taking a sample of all the theoretically possible values.  The question arises – how confident can you be that the data points you happened to sample are in fact representative of the underlying “truth” (whatever that is)?  Do you want to be 90% certain that they are representative (i.e. you will accept a 10% chance that they are not representative), or 95% certain, or 99% certain?  This is where Confidence Intervals can help.

A Confidence Interval is represented by a pair of lines, one on each side of the trend line.  The detailed mathematical description of a confidence interval is well outside the scope of this article.  However, put simply, it shows how sure you can be that the “real” trend falls within a given range, given the finite sample of measurements that you have necessarily made. When you ask for confidence lines to be drawn around a trend, you need to specify the “confidence level”, which mathematically must be under 100% - you can never be completely confident, based on just a sample.  As mentioned, common values of confidence level are 90%, 95% and 99%.

Here is a practical example of a projected trend, with confidence intervals set at the 95% level.  The data comes from a server supporting an application that was taking on an increased number of users over a period of a few months.  The objective was to predict the likely CPU utilization during an average hour at the beginning of December, so that any required upgrade could be planned in advance of the Christmas holiday season.

A simple linear trend, based on one calendar months’ worth of data, leads to a predicted average hourly CPU utilization of 61.4% on 1st December

Look at the pair of confidence lines. They are diverging rather rapidly.  Generally this means that there aren’t enough data points for a reliable prediction.  On the evidence available, you can be “confident” that the projected average CPU utilization on 1st December will be somewhere between about 35% and about 90%.  This is a pretty wide range, which you could probably have guessed without any need for statistical support.

To see what happened at the end of the following month join me again on Friday



Andy Mardo
Product Manager

Monday 7 March 2011

Plan your future infrastructure requirements in terms the business understands.

Managing day-to-day IT operations can be stressful.

Why is our e-commerce site slowing down at 11:00 pm every Thursday? How will I support the roll-out of a new service without adding more hardware?
If you had the analytics needed you’d be able to accurately isolate the problem and make an informed business decision.
Every IT shop has a dashboard of information that shows what’s happening at any given time.
Most of you will have a selection of tools but with each one only able to handle a specific task, or conduct random samplings, it’s difficult to get a complete snapshot of what’s really happening.

Capacity planning is an important step in forecasting how IT can support the growing needs of the business. One of the challenges that you have is simply knowing which servers and components serve which applications to begin with and let’s face it, without this knowledge it‘s difficult to go through a proper capacity planning process.

Our recent link up with Correlsense means that you can trace transactions and their paths through the datacenter. By feeding transaction contextual data into Metron’s Athene modelling software this data can then become the basis for addressing all of the capacity questions faced by an organization.  It means you can predict the effect of change at a granularity that means something in terms of business functions, for example, what if we get more sales, what if we merge with ACME Corp?
This provides you with a valuable insight into the performance of your enterprise data center infrastructure never previously available, and gives you a sound basis from which to plan your future infrastructure requirements in terms the business understands.

Join us this Thursday for a free webinar to look at how you can achieve this, register now http://www.metron-athene.com/training/webinars/index.html


Andrew Smith 
Chief Sales & Marketing Officer

Friday 4 March 2011

vSphere Capacity Management



A while ago I did a webinar looking at the release of VI4 (vSphere).
VMware introduced new features such as vApps, vCenter Linked Mode and vNetwork distributed Switch whilst also improving the enterprise scalability.

I delved into these new features and enhancements and what affect they would have on Capacity and Performance within a VI4 environment.
I took a look at Storage Capacity features such as Storage vMotion, Thin Provisioning and VMFS Grow, pointed out key performance metrics and reviewed enhancements to VMware Tools as well as revisiting old favourites such as CPU Ready Time, Ballooning, Swapping and Transparent Page Sharing.

So if you’d like to understand the features and enhancements of VI4, their benefits and how they can help manage capacity more efficiently, what performance metrics should be monitored and why? Or decide about Memory over-allocation. Can you do it? Should you do it? Join me, any time that suits you, by downloading my recording


I’m always wanting to update the content of my sessions and share both good experiences and bad, tips and techniques that will help us all manage VMware within the  community at large. 

Please feel free to offer any feedback you have on performance aspects of vSphere.




Jamie Baker
Consultant

Wednesday 2 March 2011

Why you need to maintain effective control over capacity.


As the enabling technology for Cloud computing and Green IT initiatives, virtualization is becoming the dominant IT implementation method.  Having emerged due to the need to control server sprawl, expert analysts such as Thomas Bittman at Gartner are already warning of the dangers of ‘virtual server sprawl’. 

The technology itself doesn’t help as it’s so easy to change resources within a virtual environment.
To realize the full benefits of virtualization though you need to maintain effective control over capacity.
Implementations need to be closely aligned with business requirements to achieve this and you need to avoid the temptation of over-provisioning solutions, because it’s so easy to do so. 
CFO’s are unlikely to countenance such activity in the future, having been bitten once already.  

Capacity, along with all the traditional ITIL® processes, such as Change Management, need to be effectively established. 
Moving to the Cloud can, to a certain degree, shift some of your responsibilities but it is worth remembering that in a Cloud world the majority of ITIL® processes remain the responsibility of the consumer, i.e. the buying company, typically your IT department, if you’re buying Infrastructure as a Service (IaaS) or Platform as a Service (PaaS). 
It’s only if you’re buying Software as a Service (SaaS) that capacity management responsibility passes to the supplier but it’s still your responsibility to ensure that the service levels offered are satisfactory from a capacity perspective and more importantly, that they are being met.
Traditional capacity planning techniques allied to some new considerations are the way to go.

If you missed our ‘Cloud Computing and Capacity Management’ webinar last month keep checking our website as some new dates will be coming soon....



Andrew Smith
Chief Sales & Marketing Officer