Wednesday, 21 September 2016

Is it worth considering SaaS for Capacity Management? (3 of 3)

I left you with the question on Monday of 'Where does this leave you?' and I guess I have raised as many questions as I have answered. 

This is no surprise as whether or not SaaS is the best model for your Capacity Management depends on your organization and your needs.  The questions are really the same as any outsourcing decision: who will do what and will it be worth the money?.

There are of course some questions that might be a simple ‘yes’ or ‘no’ answer for your organization.  If your security policies are such that any capacity data will not be allowed to be transferred out of the business, SaaS is not for you.  Understanding and getting answers to these questions first will save a lot of wasted time and effort considering options.
My concern would always be over the intangible area of knowledge.  While I can see a third party being able to meet a regular, standard, repeated reporting requirement, much of the value of Capacity Management is beyond this.  Whenever I visit our clients I am always impressed at the breadth of knowledge the Capacity Management team has across both IT and the business.  It is taking this knowledge and combining it with the raw information produced from technical capacity data that is the source of true Capacity Management value to me.  For this reason alone I can see the value in passing the logistics of Capacity Management to a SaaS provider.  Let them have the delights of keeping software up to date, ensuring there is adequate hardware resource at any time and so on. I would then usually prefer to have the intellectual capital, the use and application of that data allied to knowledge of my own business, kept in-house.  Replacing it will take time and money.

That is just one man’s opinion however.  Your organization might benefit from an expert Capacity Management business managing their capacity issues from cradle to grave, raw data to information to support your business decisions.  In this instance I would always recommend a specialist Capacity Management company with experience to bring.  Their Capacity Management knowledge and experience will be a lot harder to find than a business that can simply provide and maintain infrastructure and software for you.

If there is one area of conclusion therefore, it would be around the issue of expertise.  Whether your Capacity Management is on premise or SaaS, someone needs specialist Capacity Management knowledge and the ability to apply that knowledge to issues unique to your organization.   Whatever route you take, make sure that experience is on your payroll, or that of your provider.
If you don’t, all you might get is data.

Take a look at our SaaS Capacity Management

Andrew Smith
Chief Executive Officer

Monday, 19 September 2016

Is it worth considering SaaS for Capacity Management? (2 of 3)

As I said on Friday it's not always about the cash.
Running software in-house means more than just buying a product, then sitting back and watching it do your Capacity Management on its own.  If only the world were that simple!  Other factors are needed in place and working well to make Capacity Management successful.

Resources are needed to run in-house software.  These include machine resources such as servers, disks and network to run the chosen or developed products.  Typically today, time is needed from many teams within an IT infrastructure division to get a product up and running.  This can include desktop support, systems administrators, network support, database administrators and more.  A good installation will need Capacity Management to be integrated at a process level with other functions such as Change, Incident and Problem Management.  Getting all the right resources in place to implement and run a product, with effective processes to make sure it runs to the benefit of the business, can be a complex and time consuming task.  It all adds to the cost of that product and the risk of it taking too long see a return on investment.  The costs and involvement of resources and people doesn’t cease with implementation.  On-going support and resources, both IT infrastructure and people, are needed to keep a product delivering what it is intended to deliver.  Done well, it presents an excellent technical solution.  You have a Capacity Management product picked specifically to suit your needs.  You have a team of people, equipped with the right resources, to make the function a successful contributor to the business.

SaaS for your Capacity Management offers benefits that contrast with the software ownership model.  By paying for a service hosted by another supplier in the Cloud, you move so many of those costs on to your supplier.  Which will be the right model for you then becomes an equation. One has to compare the cost for providing those in-house resources against the charges from your SaaS supplier, assuming the end product from the two is the same.

Probably the most common assumption is that you retain the skilled Capacity Management personnel in-house. They can apply their knowledge of your business and its needs to the Capacity Management data managed by the SaaS provider.  Essentially this model transfers the administrative costs of providing a Capacity Management service to the provider.  It becomes their job to ensure there are sufficient processing resources to store and manage your capacity data.  It is their job to make sure the environment in which the software is running and the software itself, with any supporting products, is up to date, maintained and managed effectively.   The costs of providing the infrastructure are passed to the SaaS provider.  This leaves you with zero cost for creating and maintaining your Capacity Management infrastructure and zero cost on other teams within your organization to support it.

Costs are rarely absolutely zero however.  There can be many variations on this basic scenario and each organization will need to decide what suits them best if outsourcing to the Cloud.  The key is usually to define the boundaries where responsibilities meet.  If capacity data is to be captured within your organization and sent to your SaaS provider in the Cloud, one needs to define who has the responsibility for data capture.  There will be agents or agentless software to implement and manage to enable this.  There is the software and network issue of how that data gets to the SaaS provider.  Someone needs to own each part of the process and bear the cost of providing that part of the service.  Who is best placed to do that will vary with each set of needs and resources available.  All this assumes of course that your own in-house security is happy with whatever mechanism is used to pass data outside your organization, and that it’s management outside your organization meets your business standards.

Another key decision in terms of the division of responsibility is who will do the ‘value add’ Capacity Management work.  If your decision to use SaaS is purely to move the cost of maintaining an in-house infrastructure on to someone else, then you will be retaining skilled Capacity Management staff.  They will need access to the Capacity Management software in the Cloud.  Their usage of it will be as if it was an in-house owned and managed product.  They will take the raw data and transform it into the information the business requires.  This can be an efficient and effective model.  Your experienced Capacity Management team are able to leverage their specialist skills to best effect, not lose time doing work that can be better and more cost effectively done by others. 

An alternative is for your SaaS provider to carry out some or all of the skilled Capacity Management work as well as maintaining the environment.  This could provide cheaper labor costs for your organization.  On the flip side, it means you no longer retain that Capacity Management knowledge in-house.  Both its formal and informal value to your business is lost.  In this scenario, what is delivered by the outside organization tends to be highly defined and variations usually mean additional cost.  Aside from the financial concerns, some Capacity Management information cannot be effectively delivered without intimate knowledge of your own organization.  There is always the chance that something is lost if information is delivered by those who don’t have this level of interaction within your business.  It is of course a decision that needs to be right from the start.  Once you lose staff with experience, that knowledge is often gone for good. 

So where does this leave you? I'll conclude by taking a look at this on Wednesday.

Andrew Smith
Chief Executive Office

Friday, 16 September 2016

Is it worth considering SaaS for Capacity Management? (1 of 3)

SaaS and software purchase

Many people seem to be talking about Software as a Service (SaaS) for many things.  Within my own business we use it for a variety of functions: email, CRM, Help Desk, application lifecycle management – to name just a few.  We like the model and have bought into it where we feel it is the best approach.  Small businesses like ours take pride in our flexibility.  SaaS fits in with this ethos.  Money is always important to a small business.  To have moved so much functionality into Cloud based services, we must feel the cost is good.

As a Capacity Management software and services vendor, we now regularly get asked if we will provide our athene® software in SaaS mode.  Many more people have asked, than have gone forward to implementation.  This got me wondering if and why SaaS is the right delivery mechanism for Capacity Management. 
The comparison I am making is with ownership and running software, third party or developed in-house, on premise.  There is a halfway house, managed services, where the software is installed and run at your premises, but managed by a third party.  I’ll leave the managed services comparison out of things for now, as that is by far the least requested and least implemented model in my experience.

Clearly the most common model for running a capacity management function in the past has been on premise software, either developed in-house or purchased from a third party.  The relevant merits and benefits of those two approaches can be considered another time.  For now, I’ll compare having Capacity Management run as SaaS with both.  SaaS is the new kid on the block, and presumably needs to offer something distinctly better than on premise software if it is to tempt people out of their organization and into the Cloud.


The main driver for many decisions is cost, and SaaS offers a very different cost model compared to owning software.  It often boils down to a difference between ‘capex’ and ‘opex’, capital expenses and operating expenses.  Is there a cost saving for one compared to the other, or are they just different ways of accounting for money?  Although this can vary by country, the main financial difference between the two can be how they are treated from a tax perspective.  For SaaS, in many tax regimes, 100% of the software charge is tax allowable each year, against a much smaller proportion for purchased software.  For this reason alone, your CFO might prefer SaaS.

There are other related cash benefits.  From a CFO’s perspective, predictable but flexible cash flow is preferred. SaaS offers these things.  Costs are typically flat over whatever period you pay.  After some initial set up charges, payments are often made for shorter periods than software purchase, for example monthly or quarterly.  This contrasts with software purchase, typified by a high initial one time cost and an annual maintenance payment.  Unless a payment schedule is agreed with your supplier, this means a higher upfront cash outlay than for SaaS, irrespective of how the business internally accounts for that expenditure.  Being tied in to annual maintenance paid in advance can mean a business feels there is no flexibility with software ownership.  Once the payment is made, getting any cash back is unlikely.
SaaS offers the promise of being able to flex resources, and therefore expenditure, on much smaller time scales, e.g. monthly, in response to demand.
A final area where SaaS might appeal to your CFO is the speed of the return on investment compared to traditional software ownership.  That big upfront payment for software means you have to use it for some time before the benefits you receive from using it outweigh the money spent.  If you have less money spent upfront and a small monthly payment with SaaS, you cross that line much quicker.  SaaS also offers the promise of being able to downscale your Capacity Management to fit budgets and ensure it provides return on the money spent.  Software purchase means that once the money is spent, it is spent.  If the return on investment isn’t as much as expected, the expenditure side of the equation cannot be adjusted.

It’s not always about the cash…

Running software in-house means more than just buying a product, then sitting back and watching it do your Capacity Management on its own.  If only the world were that simple!  Other factors are needed in place and working well to make Capacity Management successful. I'll deal with this on Monday.
In the meantime take a look at athene®SaaS

Andrew Smith
Chief Executive Officer

Monday, 12 September 2016

How to monitor and manage disk- Windows Server Capacity Management 101 (12 of 12)

There are two main points to monitor on disk, which are occupancy and performance.

Occupancy – use Free Space Ratio%, this shows you how much space you have left on the disk.

Performance - to measure Performance we use average Response time of reads and average response time of writes.

How to monitor

Thresholds - setting a threshold for disk occupancy is dependent on how quickly you can get additional disk space and how quickly disk space is filling up, but a good rule of thumb is 70%warning and 80% Alarm.

Trends - very important when it comes to disk occupancy as it can show very far in advance when you are going to run out of disk space.

Reportsautomate reports

This is a good example of a disk slowly filling up, I could trend on this and easily get the date when I am going to run out of disk space. Having this information is very important in ensuring that there is no down time for any of my important applications. 

I hope my series has been informative and to summarize.

  • Capacity management is about ensuring that there is enough IT resource at all times.
  • Windows systems are under-utilized because of mistrust in their reliability.
  • Virtualization has helped make windows systems more utilized but not completely solved the problem
  • It's important to balance the cost of the service to the benefit
  • When managing windows systems look at CPU, Memory and Disk.

Find more white papers in our Resources section, sign up to our Community for access

Josh Worth

Friday, 9 September 2016

Monitoring and managing memory - Windows Server Capacity Management 101(11 of 12)

When monitoring and managing memory in Windows there are normally a couple of questions that people always ask, so I thought I'd deal with these.

How full can I run memory?

I find that a good rule of thumb is 90%.

How do I know when we're starting to run out of memory?

There are a number of indicators that can alert you to the fact that there are problems with memory and these are:

·       Page faults increase

·       Soft faults occur first (move things around in memory)

·       Then hard faults (write pages to disk)

·       Increases reads from and writes to page file (hard faults)

·       Reads from image files can increase

·       Eventually the system will stop responding – or just stop

In my final blog on Monday I'll be looking at how to monitor and manage disk and summarizing.
Don't miss out on our next webinar which takes place next Wednesday ' We’ve got to improve the bottom line! How Business Capacity Management can help to bring profitability to the business.'

Josh Worth

Wednesday, 7 September 2016

Memory, what to monitor - Windows Server Capacity Management 101(10 of 12)

Today we’ll look at memory and what you should be monitoring.

Memory utilization of whole system- if need be look at process working set sizes to see who’s the “culprit”, this will show you which process is using the most memory and is a good way to detect memory leaks. A good rule of thumb for memory utilization is to have at least 10% left, this is to prevent excess paging which massively hurts performance.

Page file usage% - if this is high it means that you are regularly running out of memory and windows is having to use the page files.

Memory leaks - when an application dynamically allocates memory, and does not free that memory when it is finished using it, that program has a memory leak. The memory is not being used by the application anymore, but it cannot be used by the system or any other program either. Memory leaks add up over time, and if they are not cleaned up, the system eventually runs out of memory.

How to monitor
Thresholds – when setting a threshold a good place to start is 80% warning and 90% alarm, remember if you are seeing performance issues before hitting the threshold then the threshold should be adjusted. If constantly breached, reset the value or look for memory leak.

                    Memory Utilization report, example

Above is a good example of a memory leak, you can see that memory utilization is slowly creeping up then I restart the machine it drops down and then starts to creep up again.
I'll share some best practice recommendations for monitoring and managing memory on Friday.

Josh Worth

Monday, 5 September 2016

How to monitor CPU - Windows Server Capacity Management 101(9 of 12)

As promised today we'll be looking at how to monitor CPU.


When dealing with thresholds there is no one size fits all but a good rule of thumb is 70% for a warning and 85% for an alarm these can and should be tweaked when you have a better idea of performance thresholds for your CPU.

Additionally it is good to have a thresholds in place for when a CPU is being under-utilized maybe threshold for 20% and 10% this lets you know which machine could be pushed harder.


When setting up a trend, you have to remember the longer the trend the less reliable it is. A good rule of thumb for trend is 3 months, as this gives a reasonably reliable trend and also lets you know within time to make a hardware change.


CPU Total Utilization Estd% - Report Example

Above is an example of an Estimated CPU core busy over a month for my computer with a trend going forward 1 month, you can see quickly that the trend line is going down. This kind of chart is very simple to create with a capacity management tool like athene®.

On Wednesday I'll be dealing with Memory and how to monitor this. Don't forget to take a look at our workshops, there are some great ones coming up soon

Josh Worth

Thursday, 1 September 2016

How to monitor and manage CPU - Windows Server Capacity Management 101(8 of 12)

Hyper threading is splitting a single CPU core into a two logical processors, each of these processors can execute a separate piece of work. You will see one thread being the dominate thread and one processing when the other is stalled. There is some trade off with hyper-threading as it take time for the CPU to switch between threads, Some work will fit well with this such as multiple threads of light-weight work, and more heavy work that needs the whole power of a core to get through could work slower with hyper-threading.

Depending on the type of work Hyper-threading is not always beneficial, sometimes it is better to have cores not have hyper-threading into multiple threads as the jumping between threads can lower the throughput.

On Monday I'll take a closer look at Thresholds and Trends. In the meantime why not take our Capacity Management Maturity Survey and get your free 20 page report.

Josh Worth