Friday, 22 July 2011

What can young sys admins learn from CMG-er’s, and vice versa?

CMG (Computer Measurement Group) remains the premier forum for sharing capacity management experience and the best form of cost-effective capacity management training.  Metron is proud to have been actively involved with CMG for so many years.  But perhaps there is an issue in that phrase: ‘so many years’.  Whenever I am at a CMG event, the average age of those involved is high, the number of people with grey hair or very little hair equally so.  Age tends to bring experience, and hopefully some wisdom with it, so the more senior nature of CMG should have value.

Is there an issue then, with CMG having an aging membership?  Certainly when I was first involved many years ago, many of the same people were involved as are working with CMG now, but they were a lot younger.  They had less experience then, perhaps less wisdom, but still possessed a huge pool of knowledge about capacity management. 
The question that came to my mind was whether or not this pool of knowledge is relevant to today’s young sys admins looking after more recent technologies such as VMware?  It came to my mind when playing 5-a-side soccer for the first time in 30 years, watching a colleague’s 16 year old son running around like a whirling dervish as I gasped for breath, lungs close to bursting.  I was the CMG-er, he was the VMware sys admin.

An hour later, groin strained, lungs still fit to burst, various other aches and pains fast appearing, I realized I hadn’t disgraced myself.  Sure, he could run further and faster, but I’d known enough to anticipate, see moves developing and get myself in the right place at the right time to contribute.
I see it as the same for aging CMG-er’s like myself.  The capacity management game might have changed with successive technologies.  Applications can be rolled out faster, technology is more flexible and can be changed more quickly.  Some things remain the same however. 

Hardware still costs money and money can be wasted – look at how this happened with distributed technology and server sprawl over the last 10+ years.
Poor capacity management can still lead to service level problems. So much of what we have learnt over the last 30+ years is still of value to those who have come into the industry in more recent times – but do they listen?  Do we communicate this in the right way?

The reverse is true.  Still at CMG I see people keen to hang on to doing things ‘the old way’, saying that we did this on the mainframe 30 years ago.  Sometimes this is true, but sometimes it’s not.  There might have been virtual technology on the mainframe for a considerable period of time, but it’s configurability and it’s manageability were very different to what’s available to the new generation with VMware, Xen and UNIX virtualization. 
The capabilities of each new technology should teach us something new about how to manage capacity.

So, I could save my colleague’s young son some running around by teaching him some of the things about soccer I have never forgotten from younger days.  Likewise he can teach me something about how to keep up with the faster pace of the game today.
What knowledge do you think that experienced capacity management people, archetypal CMG-er’s, can pass on to the speed freaks of modern technologies?  How does this learning need to be presented to get a hearing?

Likewise, what should the old school be listening to from the new generation?  How can we advance our capacity management knowledge by better understanding the perspective of each new generation of sys admins?
I am old (and tired, and aching from the soccer) but hopefully never too old to learn, so it would be great to hear your opinions/comments.


Andrew Smith
Chief Sales & Marketing Officer

Wednesday, 20 July 2011

What’s going to be the next “Big Thing” in the IT industry?

Today it’s Cloud Computing, but what will it be tomorrow? 

We continually strive to make the use of our IT resources as efficient as possible.  Often in the good times, it seems as though we roll out the new technology in the guise of being more efficient.  It’s the latest, so it must be the best….or at least better….or….  In harder times like the last few years, we don’t necessarily have that luxury – less money means the new technology only comes in if the payback period is short.  Not so long ago I would see people justify capacity management software over a three year payback period.  Last year I worked with one who had to show payback in three months.

Of course, Capacity Management is critical to gaining that efficiency you seek whether you’re rolling out new technology or making the most of what you already have.  Any unplanned change is likely to have cost implications that are higher than they would be if you’d used good capacity management practice to plan ahead.

Server sprawl was replaced with virtual server sprawl in organizations that didn’t capacity plan the migration.  Cloud could reduce that virtual server sprawl, but unplanned usage of Cloud resources will carry a premium cost and erode cost benefits of the Cloud over time.

The best answer is always to look at what the business needs.  From Metron’s capacity management perspective, one of the tools in our arsenal to achieve this is Business Transaction Management.  This allows the Capacity Manager to look at their environment from the traditional view, silo by silo, piece by piece, but also look at it from the End-User perspective.  For example, being able to map a user transaction through its life cycle allows the Capacity Manager to look at both the breadth and depth of the environment.  Issues and requirements are seen from a user and business perspective.  The effect of differing IT resources and technologies on these are easily seen.

Our technology for achieving this takes Metron’s Athene in conjunction with Correlsense’s SharePath to provide that breadth and depth.  This combination lets you see how much good tuning existing infrastructure can do for the business, how well the business can be supported for a given cost with new technology and much more.  We’re excited about the opportunities afforded by using these two products in tandem to move capacity management forward from a technical task to a value-add business process.

The bottom line is whatever the next “Big Thing” is that comes along, the Capacity Manager will be able to effectively manage their IT resources. 

Talk to us about how Athene + SharePath can help you get more from what you have, and make the most of wherever the technology takes us next.


 Charles Johnson
Principal Consultant


Monday, 18 July 2011

Capacity management software is a compromise. The definition of a compromise: nobody gets what they want

Imagine a fair system to vote for painting your office.  There are four choices: red, blue, green and yellow.  Three people share the office and rate their preferences as follows:

Preference
Color
Person 1
Person 2
Person 3
Total
1st
Red
1
3
4
8
2nd
Blue
3
4
1
8
3rd
Green
4
1
3
8
4th
Yellow
2
2
2
6

Choices are rated accordingly and totalled. The lowest total is on average the most preferred option, so the winning color is announced: yellow.  All three people groan: no one wanted a yellow office!
That’s the nature of a compromise – it can mean that no one actually gets what they want. 

Any packaged software is a compromise.  No packaged software meets every requirement of every user.  Packaged software by its nature attempts to provide the features that meet the needs of a particular segment of the market for that product.  This could be the most numerous, the most vociferous or a host of other criteria.  The underlying fact however is that not everyone will get everything that they want.
Capacity management software is a form of packaged software.  Vendors do their best to meet as many needs of their clients as they can, but it is impossible for any one product to meet all of a client’s needs.  It can be a simple thing: one client looks at memory first, another at CPU first.  Each would like their most common selection to be at the top of the menu.  It can be more complicated: you gather your data from IBM’s Tivoli for all systems and pass that data to your capacity solution, except for Application X which is not supported by Tivoli.  For Application X, you need another data collection method.

IT infrastructure gets ever more complex.  We had mainframes, we went distributed, virtualized chunks of our world and now we are putting some of it into the Cloud.  Having a product that can manage the capacity of all of that, out of the box, gets ever less likely.  Constraints on implementing IT infrastructure support software such as capacity management get ever more stringent, for example the need to meet tighter security obligations.
Having solutions for tasks such as capacity management that allow you to build on core facilities and tailor the package to meet your own needs are now ever more important. 

We have a very flexible solution here at Metron that can be used out of the box to meet the majority of your needs and constraints.  For those awkward areas that need a custom solution however, we have a variety of ways in which we can adapt our core facilities to ensure your problem children are cared for.

We’d be interested to hear about what challenges in this area you have had to overcome, whatever package you have started out with, so we can check we are on the right track.
Now, back to staring at my yellow office walls and deciding what to do next……

Andrew Smith
Chief Sales & Marketing Officer

Wednesday, 13 July 2011

Do people look at you weird when you mention Capacity Management?

Many times there is difficulty in providing the “Elevator Pitch” when describing what that term means.  It’s easier for some.  When I was a younger man, filled with post-Business School enthusiasm, remembering the advice that I should show an interest in a client or prospect’s business, I asked an officer of Air Force Global Strike Command what his part of the military actually did.  With a smile on his face he said, ‘Well, if you want a bomb dropped on someone, we’re the people to ask’.  The perfect ‘Elevator Pitch’ – less than five seconds and a perfect understanding was communicated.….and no, I never did ask.

Most Capacity Managers can’t do this for their role.  Other individuals within an organization have even more trouble than us because they don’t understand the meaning of Capacity Management or how this affects their daily jobs.  I’ve tried over the years with a variety of similes, in particular trying to relate capacity management to non-IT life:  it’s like putting a dashboard into a car (no good ignores planning/prediction); it’s like predicting where traffic snarl ups will occur on the freeway (‘Why don’t you use it to stop traffic snarl ups happening then?’).  You get the idea – I’ve never found a simile quite good enough.  I’m OK within IT: ‘it’s all about analyzing, reporting and predicting the performance of computers’ is a good start for me.  IT people understand this, but for those outside IT, you watch their eyes glaze over.  It’s when we need to communicate to those who don’t understand IT that things become difficult.

With this in mind, to help others in our organization understand what we do, it is key that we use non-technical verbiage to express how what we do affects their job.  An effective way to express what we do is to use the words  “pain” or “cost”.  It usually gets our attention if our doctor says ‘there might be a little pain’ or our offspring say ‘there could be a little cost involved’.

A statement such as “Effective Capacity Management will allow your data entry staff to process 20% more invoices per hour without any increase in cost” can be an eye opener for them.  Likewise, ‘it will remove the pain of lost business when the on-line ordering service slows down, without impacting your budget’.  If we continually look at how we present our data and how we talk to the business users, we can get more buy in from the business groups.  The ability to incorporate business data into capacity management afforded by Metron’s Athene software and our business view of transactions provided by working with Correlsense’s SharePath software helps us bridge that gap.

I challenge us to develop an elevator pitch in describing what we do and how we can affect the enterprise.  What’s your best five seconds of words explaining to someone non-IT literate what capacity management is all about.

‘Capacity Management? Well, it’s like…………………………………………………………………...’ 

Over to you.

Charles Johnson
Principal Consultant


Monday, 11 July 2011

Cloud Services Demand Flexibility from Capacity Planning Tools

I think the ability to extend the functionality of available tools to meet emerging demands is of key importance when doing Capacity Planning for a Cloud implementation.  

Capacity Planning is always a complex task.  Complexity is especially pronounced in a Cloud environment where system and workloads are mobile and dynamically scaled.  Capacity Planning in a Cloud environment requires robust tools which can be adapted to immediate and emerging needs.  Such tools must have the scope to address all the various components to be accounted for and the flexibility to acquire, analyze, and report from the data available for those components. 
The variety of systems, software, and related components that are present and potential in the Cloud mandates a great range of functionality from tools that can only be satisfied if the functionality of the tools can be modified and extended to accommodate combinations of data and data sources as they emerge.  Data acquisition and marshaling capabilities must be comprehensive and consistent to enable the expected quality of analysis and reporting.  Results and report distribution schemes must satisfy current and future requirements as well.

Generally, any given software tool for capacity management will not be able to meet all your needs.  This means that ideally any tool will enable you to extend the scope of featured data processing functionality with the availability of both pre and post data processing capabilities. For example, data may be available from data collectors which provide useable metrics but which are not directly supported by the product.  In this circumstance it is nice if you can create a script or program (user’s choice of language) which reformats the available data to match a supported data source, or even create a custom record format and database to suit your needs.  In either case, data of interest can then be successfully loaded into your capacity management solution for analysis and reporting, assuming you have a solution that is sufficiently flexible.  Similarly, featured reporting capabilities may be extended with post processing capabilities developed by or on behalf of the user.  Any good product will allow and support this.  IT is a sign of strength not weakness if a package offers pre and post processing extension capability as part of the monitoring and the reporting processes. 
In addition to data collection and reporting, the ability to extend processing beyond the released feature set provides users with convenient access points to interact with external systems and processes and extend both the functionality and scope of the tools.  For example, a threshold monitor may flag resource utilization and then, via a processing extension, an event notification may be passed to a work scheduling system for scheduling or an operations console for prompt remediation.  In complementary fashion, it’s good if your capacity solution can acquire dynamic information such as alerts and updates provided by external systems and incorporate that information into ongoing processing and reporting. 

I’ve been heavily involved in extending Metron’s Athene software in these ways to provide highly tailored solutions to meet specific client needs.  I’d be interested to know of any thoughts you have on challenges in this area, whatever the capacity solution you are using.

GE Guentzel
Pre Sales Consultant

Wednesday, 6 July 2011

I wonder what you're doing this Summer

During the summer, there’s downtime in organizations as many people are on vacation.  For some businesses this can also be a time when ‘business as usual’ offers slightly less stress and we can turn our attention to some of the background activities that enable the rest of the year to move forward successfully.

Like many people, I often see the summer as a time when I can do a bit of formal or self-training to keep my skills up to date.  As things slow down, it’s a good time to brush up on our skills in certain areas.  Sadly, formal training often has to take a back seat as work pressures these days just don’t allow for several days out of the office attending a course in a classroom. 

Some areas of learning don’t lend themselves that well to classroom-based training anyway.  I’ve always experienced problems with any non-product specific IT training I’ve had, especially capacity management training.  I find that I need to have fairly instant reinforcement of training through on-the-job usage of those skills, or quite simply, the memory fades over time and much of the training is lost. 

Capacity management particularly fits this bill for me.  You get some great tip or technique for tuning or sizing a system, but those little tips are the first things you forget as the classroom-based hands-on exercises are usually based around the larger more general issues.  Then of course there is capacity planning.  It’s fine to get taken through the theory and even a practical exercise re. building a capacity plan in the classroom.  Back at the office of course, it’s then usually several months before those predictive and business liaison skills are called on.  No matter how good the notes I scribble on to that iPad, it’s not the same as practicing the skills right after the training. 

To try to help you over these limitations of classroom-based training and use some of that summer time to good effect, Metron is presenting its first Summer School beginning July 12th 2011.  This year’s Summer School is around VMware Performance and Capacity Management.  This class is focused on giving the individual practical troubleshooting techniques to assist in monitoring their VMware environment.  There’s quite a bit of information out there and this set of classes is meant to assist in consolidating that information.  

There will be four one-hour sessions spread over an eight week elapsed period.  This means that you get the chance to go and practice the skills taught each week, before moving on to the next section of the course.  That immediate consolidation of learning aids retention of information a huge amount.  There’s the ability to replay a session if you can’t make the on-line event, which means it can fit in with even the busiest summer diaries.

From a technical perspective, the benefits in attending are:

  • Identify common performance and capacity issues with VMware
  • Implement a structured approach to resolving such issues
  • Understand and manage the interaction of key system components
  • See through issues masking the root cause of a problem and resolve that root cause

Even though the Summer School isn’t related to our own Athene software in any way, Athene clients can attend free. 
CMG users get a discount and even if you don’t qualify for either of those categories, it only costs $249 for all four sessions. 
This means the boss won’t come back from his holiday and find you’ve broken the budget.

We look forward to having you join us.  For more information, please go to:  http://metron-athene.com/training/summer_school/index.html

Charles Johnson

Principal Consultant

Monday, 4 July 2011

Confused by Capacity in the Cloud? Don't be . . .

Cloud computing comprises of the following three key services:

·         Software as a Service (SaaS)

·         Platform as a Service (PaaS)

·         Infrastructure as a Service (Iaas)

Clouds are typically categorised into External(Public),  in which External Service Providers (ESP) such as Amazon or Google charge for the service (SaaS) or resource usage (PaaS / IaaS) accordingly. 
Internal(Private), in which an organisation will host the services and Hybrid which is a mixture of the two and enables 'cloud bursts' in which services can be ported between external and internal clouds.

In order to be considered a cloud, a service must provide:

·         On demand self service

·         Resource Pooling

·         Rapid Elasticity

Therefore, most if not all External or Internal Cloud Computing physical architecture is underpinned by a virtualization technology, such as VMware vSphere, Microsoft's Hyper-V or Xen.  These technologies provide the above functionality and allow the cloud service providers to 'carve up' their resources as necessary to satisfy the demand of the consumer.

Only the SaaS (Software as a Service) model of cloud computing moves the responsibility for capacity planning to your supplier. If you're buying a platform, infrastructure or an application as a service, it is down to you to make sure you ask for the right resources.
Buy too much and you are perpetuating the over-provisioning mistakes of the distributed era. Buy too little and you bounce from performance crisis to performance crisis and pay the costs of short term fixes. Traditional capacity planning techniques allied to some new considerations are the way to make sure you only pay for what you use.

So how to do we capacity manage an internal cloud environment?  Register for our seminar at Vinopolis on 20th September in London to find out . . . .. . . . http://www.metron-athene.com/training/seminars/index.html

Jamie Baker
Principal Consultant











Friday, 1 July 2011

Summer School, cost effective web-based training on managing your VMware environment.

Work schedules and the need to be ‘on the job’ have made it increasingly difficult for people to justify attendance at training courses. In recognition of this we’ve come up with a short sharp high intensity training class in performance and capacity management of VMware environments. The aim is to try to give you as much of the ‘classroom experience’ as we can, without the down side of attending a formal scheduled class.

The training is delivered as four one hour web-based training sessions on July12 and 26 and August 9 and 23. Meetings, meeting, meetings and other job commitments mean you can’t always make it to a session.  If work schedules mean you have to skip a session don’t worry - you can catch up by listening to the recording at any time up to two weeks after the class.
What makes us uniquely qualified to deliver this course?

We’ve worked with VMware as both a technology and consulting partner, including provision of VMware branded consulting on a sub-contract basis. We’ve provided our ITIL accredited Capacity Management training course for VMware capacity management staff and we’ve also worked with VMware to create the Capacity Management Service Acceleration Kit for use by VMware re-sellers worldwide.  Add to that several years providing webinars for the public and free training for our Athene users across the web, and we have a good skill set in delivering information this way.
You’ll be taught to identify headroom for additional VMs in a Cluster, Dynamic Resource Scheduling, setting up Resource Pools, VMware KPI’s, identifying Idle and Oversized VMs, the best way to present reports to end-users and much more. You will be able to identify common performance and capacity issues with VMware and implement a structured approach to resolving them. Our aim is to help you understand and manage the interaction of key system components and see through issues masking the root cause of a problem and resolve that root cause.

So who should come along to our Summer School? Well our class has been designed for anyone who has an interest in ensuring good performance of their VMware systems and anyone who has ever encountered a performance problem with VMware.
You won’t need more than a basic understanding of VMware concepts and architecture to derive benefit from the class and all concepts, advice and recommendation are ‘product agnostic’ and don’t rely on any non-standard VMware or third party tools being available.

Our Summer School suits both VMware specific staff and those who need to know more to help in their own sphere of influence, even though they’re not specifically VMware technicians.
You’ll probably get most from the course  if you’re working in areas such as VMware system administration, performance management, capacity management, infrastructure management, network administration, database administration and application support

So if you’d like to enhance your ability to identify VMware performance issues and tune your environment to provide better levels of service why not sign up now.
http://www.metron-athene.com/training/summer_school/index.html

We’re always looking to enhance the ways in which we deliver information to people – feel free to respond with your thoughts on the validity of this on-line training approach.




Andrew Smith
Chief Sales & Marketing Officer









.