The up.time IT Systems Management Blog

Is your Capacity Planning Evolving to Meet Business Demand?

May 11th, 2012 pdidaskalou

 

As an IT systems management vendor, we get fired up about new technologies including the latest buzz around virtualized capacity, automation and cloud. We respond by building slick tools, dashboards and reports to help solve capacity problems. I believe that’s what (we) systems management providers are supposed to be doing, helping you solve problems. <shamelessplug> Reducing the complexity of capacity planning and management is something we do really well around here at uptime software! </shamelessplug>

capacity planning

Capacity Management is all about evolving IT Operations.

But what about the capacity planning function itself? Does it not need to evolve along with these new deployment technologies? Do current capacity planning functions contribute value to the business by helping them scale to meet demand?

Virtualization, automation and cloud technologies give IT execs more options than ever before in how services will be delivered to the business, but do their current capacity planning processes reflect this same evolution in technologies? For most the answer is still likely “no”.  Most IT organizations still seem to perform capacity planning at the individual component level (server, network, SANs) which does not represent the true capacity requirements of their global facilities and infrastructure resources. The good news here is that you CAN evolve and turn this situation around.

Planning and managing IT capacity at a macro level is critical to delivering cost-efficient and reliable business services in a time frame the business expects. The good news is that today’s virtualization and automation technologies allow flexibility and new cost alternatives so IT execs can choose from a myriad of platforms to run applications and services on. The bad news is that these new virtual and cloud based resources are certainly not free and without new capacity planning processes, the benefits of easy procurement and instant provisioning can quickly turn into over-allocation and cost overrun nightmares.

  • So the message is clear: IT executives need new and more effective capacity planning processes in order to really take advantage of new technologies by optimizing the placement of applications according to criteria such as service level and cost. In addition, capacity planning software and tools can help teams be more effective.

One tactic you might consider as a start is to elevate your capacity planning team. Get it out of the “back room” of IT operations and make it a strategic function. Yes, remove it completely from IT operations and centralize it as a corporate IT function that reports directly to the CIO. This will send an important message to your organization and capacity management will begin to evolve and operate decentralized from technology support groups, such as network, server and storage.

capacity planning software

But Rome wasn’t built in a day….

Tags: , , , , , , , ,
Posted in automation, Capacity Planning, cloud, IT Management, IT operations, management, sla, Systems Management, Virtualization | No Comments »

ShareThis

Capacity Planning: Do you Know your Virtualized Environment?

May 9th, 2012 Patrick Lui
Remember, the key to capacity planning is knowledge. You cannot manage something you don’t know.  And to stay on that point, since VMware is probably the most popular virtualization technology available, I want to elaborate on VMware capacity planning.

 

up.time communicates with VMware’s vCenter to get all sorts of data and metrics.  For capacity management, we tie all that information neatly into easily digestible reports.

One of these reports is the vSphere Workload Report.  It enables you to visually see the resource usage at the data center, cluster and ESX server level as well as resource pools, vApps and the virtual machines.  With this knowledge in hand, you can easily determine if you are maxing out your VMware environment or if there’s room to better utilize your resources.

 

Quite a lot has been said about our capabilities on VMware but that is not the only virtualization technology we work with.  Another one that we have coverage for is IBM’s LPARs.  The LPAR Workload Report shows the CPU, Memory, Network I/O & Disk I/O for all the LPARs on a pSeries server. The graphs are stacked so not only can you see how the LPARs compare to each other, you also gain insight into the overall workload on your pSeries server.  This empowers you to accurately adjust the CPU entitlements of the LPARs and keep track of the overall workload over time!

Capacity planning might sound like a difficult task to tackle but through my last few posts, I hope I have shed some light on how up.time can assist in fulfilling your capacity management needs.  If you haven’t already done so, download a free trial of up.time and see what it can do for you!

- Patrick

Tags: , , , , , , ,
Posted in Capacity Planning, IT Management, monitoring, Systems Management, Virtualization, virtualization vmware, VMware | No Comments »

ShareThis

up.time VMware vSphere 5 Support

April 27th, 2012 Dave Leith

You asked for it, our upcoming release of up.time 7.0 will fully support VMware vSphere 5! uptime software inc. has been a long time VMware partner, launching our first VMware monitoring solution in the early days of ESX in 2006. Since then we have released several major evolutions of our virtualization monitoring platform and are continuing to build on our relationship by fully supporting the latest vSphere edition in up.time 7.0. up.time 6.0 already fully supports vSphere 4.x, with all of the powerful features you are used to continuing to be available:

Smart VMware Monitoring

  • vSync Dynamic Discovery: Keep your monitoring inventory in lock step with any changes to your vSphere environment. The second a new VM is spun up or hosts are moved around the datacenter, up.time’s inventory will update itself instantly so that you know you don’t have any monitoring blind spots in your environment. Monitor your whole vSphere environment agentlessly through your vCenter installation.
  • Power Awareness Intelligence: Power state awareness dashboards bring real time power status information into your global view. Take control of power state changes by alerting your administrators when critical systems are powered down. 
  • VM Sprawl Control: Automatic notifications of new VMs ensure they are compliant with your IT policy. Review weekly sprawl reports to target zombie or over allocated VMs  so you can free up precious disk & memory space to the VMs that really need it.

Deep VMware Capacity Management

  • Easy Capacity Bottleneck Troubleshooting: Instantly find your bottlenecks and drill into the root cause.
  • Global Capacity Reports: How much capacity do I have? How much am I using? How much am I wasting? up.time helps you answer all of these questions so that you can reclaim valuable resources and proactively avoid embarrassing capacity outages.

In addition to all of the other great benefits of up.time, our “single pane of glass” dashboards brings together your virtual, physical and cloud environments into one monitoring toolset.

If you haven’t had a chance to try up.time yet, you can download a free trial from the VMware Solution Exchange or our website and be up and running in 15 minutes.

Dave.

Tags: , , , , ,
Posted in cloud, Virtualization, virtualization vmware, VMware, VMware Monitoring | 2 Comments »

ShareThis

Cradle To The Grave – Virtualization Capacity Management

April 19th, 2012 Patrick Lui

 

Virtualization is not a new technology.

Virtualize Better
Companies of all sizes have utilized virtualization in a number of ways for a long time, even back to UNIX (yes, AIX LPARs and Solaris Zones).  However, how does virtualization help in the context of capacity planning and management in today’s complex IT environment?

One of the key benefits of virtualization is its ability to provide flexibility.  If demand is surging, you can spin up new virtual machines almost instantaneously to meet it.  So, the real question then becomes what should you virtualize?  This isn’t always an easy question to answer because you need to know both how much resources/bandwidth you are currently using, you also need to predict how well the VMs will work on your target physical host.

As mentioned in an earlier post, knowledge is key to capacity planning.  In up.time, the “Server Virtualization Report” helps provide that essential knowledge by automatically identifying physical servers that are good candidates to be virtualized on your target machine.  Simply put, up.time can predict if your physical servers will play nicely in a virtualized environment.  It gives you insight into the future, so you don’t have to spend weeks going through a physical-to-virtual exercise or commit  to buying hardware before you know exactly what you need.

Server Virtualization Report

Server Virtualization Report in up.time

Companies that have taken full advantage of virtualization technologies often end up with another problem: VM sprawl.

Since it’s easy to spin up a new Virtual Machine, many users have a tendency to request VMs all the time.  The result can be a data center infested with VMs that either aren’t being fully utilized or worse yet, not used at all.  This is what we call VM sprawl and it’s a headache for companies that are virtualizing.  These VMs are costing businesses real money in resources that aren’t being used.  In other words, they are taking up capacity that could be used to achieve more important objectives in your organization.

VM Sprawl Report

The "VM Sprawl Report" in up.time

How can up.time help?  A built-in “VM Sprawl Report” identifies the extent of the sprawl across your virtual infrastructure. It shows VMs that are underused, always off, suspended and even VMs that have not been powered on recently.  It also shows trends that can tell you if sprawl is growing or being reduced (a great report if you have a project to reign in sprawl and need to track your progress).  up.time empowers you to identify and solve your VM sprawl problems.

As you can see, we have reports that help you plan and execute your virtualization initiatives from cradle to the grave.  Every company, no matter the size, can reap benefits from using up.time.  Download a free trial of up.time today and take it for a test drive!

- Patrick

Tags: , , ,
Posted in Virtualization, virtualization vmware, VMware, VMware Monitoring | No Comments »

ShareThis

New Network Monitoring – Device Dashboard Sneak Peek

April 17th, 2012 Dave Leith

Our next big release of up.time will be “up.time  7 (coming soon),” and I wanted to give you a sneak peek into some of the powerful new Network Performance Management capabilities, including deep performance analysis for network devices. Out of the box, up.time 7 will collect an array of deep performance metrics from your network devices.  All you will need is SNMP credentials and the rest of the work will be done for you by up.time. No manual setup needed for best practice performance metrics.

The new “Network Quick Snapshot” dashboard for network devices (shown below) is designed to give you an instant summary of device performance and visibility into key bottlenecks that could be impacting service levels. In this case, we’re looking at a Cisco Catalyst switch sitting in front of a Dell EqualLogic iSCSI SAN device, hosting dozens of VMware datastores. The VMs using this SAN depend on the Cisco switch being up and running with enough capacity to meet any workload demand that is thrown at it. Key metrics including latency, packet loss, in/out bandwidth, errors and discards are available instantly for your review.

In addition, potential network hot spots and problem areas are highlighted to save you time. From here, you can drill down into specific metrics for deep root-cause analysis of performance problems or compare current performance to longer term trends in capacity, for example. This ensures you stay one step ahead of any potential network capacity problems, and helps you move from a reactive IT mode to a proactive IT mode.

Don’t worry, you don’t have to watch the dashboard all day long to stay on top of your network! As usual, up.time will keep an eye on all the performance, availability and capacity metrics and alert you (email, sms, etc)  if there are any problems or potential threats that need your attention.

- Dave

 

Cisco Switch Quick Snapshot

Tags: , , ,
Posted in monitoring, Network Monitoring | No Comments »

ShareThis

Navigating IT and Business on 2 Wheels

April 13th, 2012 pdidaskalou

Ok, this won’t be a typical post. However, as the CEO of uptime software, it’s something I’d like to share with you, our customers and prospects.

Building our software company has been a passion and a source of adventure for me since we began operations in 2001. It has been a wild ride to say the least. At the same time, living and working in Toronto, Canada, has also given me the opportunity to explore, with great zeal, another passion – motorcycle riding. You see, we have access to some fantastic and scenic rides within two hours of Toronto.

I have been an avid motorcyclist for 18 years. When I’m not in the office, I try to spend most of my free, warm weather time cruising long distances on my Harley Night Train, or carving twisty roads on my Ducati 848 super sport motorcycle.

With over 238,000 kms on to two wheels (so far), I’ve had to deal with my fair share of unexpected events. They usually happen in the blink of an eye. However, learning to deal with the unexpected is a skill every CEO needs to hone. We grew uptime software from the simple idea that IT systems management should be easier. What started 11 years ago with $1,000 in seed money (and copious amounts of blood, sweat and tears) has become a multi-million dollar company that continues to grow and add new enterprise customers every day. There’s no doubt, it was a team effort to make uptime software the success it is today. However, for my role, I credit my results, in part, to the instincts I honed while riding on “the street.”

How? Flexibility and Adaptability.

My experiences on the road have taught me to change quickly, when conditions change. To shift gears and accelerate, hit the brakes, or just coast to collect more information on the landscape. Quite simply, it’s learning to expect the unexpected.

My Steering Principles for Managing Unexpected Events on the Road and in the Business:


1…

On the Bike: Know when to punch the throttle. Sometimes you have to take some risk to break away from the pack.

In the Business: Know when it’s time to ramp up your operations and teams. Sometimes that means adding sales resources, marketing budget, developers, or a combination of them all. Understanding customers’ needs and the flow of the market enables you to decide… is now the time to downshift and hit the throttle? When the right opportunity presents itself in the market, it’s important to make an aggressive move in order to serve customers better, attract new ones, and leave competition in the dust.

2…

On the Bike: Sometimes everything looks fine and you can just cruise along enjoying the scenery. However, you always need to have your eyes well out in front. That allows you to see potential changing road conditions, construction, accidents or a brutal traffic jam. If you can see far enough ahead, you give yourself (and your business) options, like getting off the main road. From there you can find an open road and keep moving forward instead of breathing others’ exhaust.

In the Business: Business is always competitive and in some cases you can end up surrounded by competitors. That doesn’t help you and it doesn’t help your customers. So…get creative, innovate, break a new trail, don’t be a follower. It’s the perfect opportunity to find new market, choose an alternate road and keep your forward momentum. Your customers and your employees will thank you for it.

3…

On the Bike: Use all four eyes. The ones above your nose and the ones you’ve installed in the back of your head. If you ride a bike, you know exactly what I mean. It’s important to be prepared for any unexpected moving objects. It can be the difference between a great ride and an ambulance ride.

In the Business: Have eyes on the market. Stay educated, network with peers, and hire smart people. If you constantly feel the pulse of the market and have a great team surrounding you, your company will be able to sense industry trends and technological shifts well ahead of the curve. In business, it’s imperative to anticipate the big changes, and get on the right side of them. Train your brain to anticipate.

Running this business and riding are my passions. Funny enough, as different as they are, I use a similar skill set when dealing with unexpected events for both. These simple yet effective tactics have increased my joy, success and ultimately my longevity in both my business venture and my motorcycle riding.

- Phil

Tags: , ,
Posted in IT Management, uptime | 1 Comment »

ShareThis

Never Run Out of Disk Space Again with Capacity Management

April 3rd, 2012 Patrick Lui

Get Better Capacity ManagementLast time, we showed how you can quickly determine if you need more hardware based on how much CPU is being used in your data center.  I can hear some of you screaming…”DISK!  What about disk??  How do I know when I will run out of disk??”

 

And you know what, you’re absolutely right.  You should be thinking about your disks as part of your capacity planning initiatives.  If you are still unsure why you should care about capacity planning, take the flood in Thailand as an example.  Hard drives have shot up in price since the flood.  If you can foresee when you will run out of disk space (and how long you can live without it), maybe you can avoid paying an arm and a leg for additional capacity when you really don’t need it yet.  How?  Look no further than your trusty up.time Monitoring Station!

IT Capacity Growth Report

See File Capacity Growth Over Time!

 

By using up.time’s File System Capacity Growth Report, you can quickly see how your file systems are filling up.  This single pane of glass report takes the specified period and calculates how the file systems grow/shrink.  For example, if you selected to report on a one-month period, the delta shown will be the rate at which the file systems are growing/shrinking for the month.  Therefore, you can arm yourself with this information and be able to plan when you will need to buy more disks.

How long until YOUR VMware disk space is all gone?

 

 

If you are using VMware, you have additional tools in your arsenal to succeed at capacity management.  up.time can calculate approximately how long it will take to fill up your VMware datastores… automagically!  It tells you, in English, if your datastore will be full in 3 months, 1 week, or whatever the case may be, so you won’t be caught off guard by running out of disk space.  What makes this even cooler, is that we do it agentlessly!  Just add your vCenter or ESX servers to up.time (which takes about 5 seconds) and you will be able to see into your VMware future with the up.time VMware datastore crystal ball!

 

Free Capacity Management Webinar: 3 Simple Steps for Total Control of your IT Capacity.

Tags: , ,
Posted in Capacity Planning, IT Management | No Comments »

ShareThis

Key to Capacity Planning is Knowledge

March 20th, 2012 Patrick Lui


Put your books way.  This is a capacity planning and capacity management quiz.  How many of these questions can you answer?

  1. How many resources are you using across your entire datacenter?
  2. What is using those resources?
  3. What is the utilization trending towards?
  4. Can you add more load to your servers?
  5. When will you run out of capacity?
  6. Will you gain much from virtualization?
  7. Do you know which servers you should virtualize?
  8. If you are already virtualizing, are you using your resources optimally?

 

Times up.  So how did you do? If you don’t have it memorized, I sure hope you have answers to the above at your fingertips, as those are some very typical questions that CIO’s would ask.  Most people view datacenters as cost centers, and for good reason.  Rising energy costs make headlines everyday and, given the slow recovery of the economy, we are constantly being asked to do more with less.  However, how can you do more with less if you’re not even sure of what you already have.

If you didn’t do so well on the test, I’ll let you cheat a little.  Download and install a trial of up.time.  For those of you who use or have heard of up.time, you might only know up.time as a monitoring tool.  However, there’s much more to the up.time solution than you may realize (and it’s already included in the tool): capacity management.  up.time has loads of features to help you perform capacity analysis.  There are many examples and success stories of how up.time can solve your capacity pains but let’s focus on one common scenario.

Example #1:

Your system administrators come to you saying they need to buy “X” number of servers because they are running out of CPU power.  You, as a diligent manager, know that the executives are never pleased with additional budget requests. So, before you put your neck on the line  for more money, you need to quickly verify what your admins said  is true. Fear not, my friend.  Simply pull up the up.time web UI and generate an “Enterprise CPU Utilization Report” for your entire datacenter in about 3 clicks.  Voila!

 

In seconds, you can easily see how much resources your servers are really using.

 

This is just one of the many examples of how you can use up.time for better, faster, and easier capacity planning.  We will be sharing more on how you can use up.time to address your capacity analysis needs over the next few weeks.  Stay tuned!

 

 

 

In fact, join us for a Free Webinar on Capacity Planning and Capacity Management:

- Patrick

 

Tags: , , , , ,
Posted in Capacity Planning, IT Management, uptime | 3 Comments »

ShareThis

Network Monitoring – New SNMP Poller Sneak Peek

March 16th, 2012 Dave Leith

 

Improved Network Performance Management has been a long running request from up.time users. Coming from primarily a server and virtualization monitoring history, network monitoring certainly hasn’t been a focus for us in the past. I’m happy to say that we’re listening, the next version of up.time’s monitoring solution is aiming to dramatically improve our network monitoring capabilities. We’ve been working on this for a few months and are very happy with how it is turning out. I’d like to give a preview of one of the first major pieces of functionality in the next release, our freshly redesigned SNMP Monitor that aims to make collection and alerting on custom SNMP data as easy as possible.

Why redesign the SNMP Monitor? The old monitor had a number of limitations that made it far too difficult to setup in a way that provided valuable results. Ease of use and time to value are very important to us and this monitor wasn’t up to the standard we, much less our users, expected. We also found that the old monitor didn’t do many of the basic things we needed it to while building our other network monitoring components. So we decided to redesign with these goals in mind:

  • Improve the ‘out of the box’ options for custom SNMP values by including a much wider array of vendor MIBs so that you don’t have to go searching for MIBs
  • Reduce the pain of managing and changing SNMP data points, this process should be as easy as any other monitor in up.time
  • Focus on flexibility, we want you to be able to gather just about any SNMP value you can think of and use it in a way that makes sense for your environment

Here’s a screen shot of the new SNMP Poller configuration page:

New SNMP Poller

Here are some basic examples of things that you can do with the new SNMP Poller that you either couldn’t, or found very difficult to, accomplish with the old SNMP Monitor. The examples below were configured in 8 minutes total then let run for a while to gather some real world performance data.

Handle Large Tabled Values: Here we’ve setup a monitor to collect ifInOctets across all interfaces on a device

Table Based Performance Metrics

Monitor Vendor Specific Info: Here is a graph of server temperature gauges collected from an SNMP enabled Dell server

Vendor Specific Info

Monitor Hardware Status: Here we see a few monitors checking power supply status and processor temperature

SNMP Monitor Status

 

Thx,

David

Tags: , ,
Posted in IT Management, monitoring, NetFlow Monitoring, Network Monitoring, Systems Management | No Comments »

ShareThis

Cloud Performance and What It Really Includes

March 15th, 2012 Alex Bewley

It’s been a while since I last wrote a blog entry, but with good reason.  We’ve been busily working away on uptimeCloud, our latest and exciting offering for cloud cost management (browse to the site at http://www.uptimecloud.com).  uptimeCloud is the simple way to manage the cost of cloud. I’ll write another blog on this subject in a little while, as there is a lot to talk about. We’ve been getting some great media and analyst reviews and if you want to know more about it, here’s a great video:

Cloud Performance Monitoring

In the interim, I wanted to talk about “cloud performance.”  So, what does this mean and what can we do to address concerns that arise?

Cloud performance can mean many things: from simple latency between an end-user and a hosting provider; to availability of the applications in the cloud;  to very complicated monitoring of highly interdependent services in the cloud.  For example, on your desktop you probably have lots of ‘cloud’ type connections that you may not be aware of, such as Evernote (for cloud-based notes), Dropbox (for cloud-based storage), MobileMe (for calendaring and contacts), Gmail, Gtalk, SalesForce (for CRM), a random mix of various hosted applications, and there can be an assortment of special business application dependencies as well.

Add in a hosted application dimension and things get even more complicated:

  • IT service dependencies (e.g. using spot instances of Amazon’s AWS for specific compute workloads, managed by a separate cluster);
  • using a service like PiCloud for millisecond billing;
  • depending on a service like Twilio for SMS message delivery;
  • putting storage in S3;
  • cloud hosting in AWS, Rackspace, GoGrid; or application hosting with force.com).

Of course, these services are dynamic and may scale up or down depending on usage.  An application can break if any single one of these components breaks.  It really makes your head spin.  If you read the fine print on most of the services, it’s all on a best-efforts basis.  So while you may have availability on certain key services, they may depend on adjacent and not quite so reliable services.

Just for fun, add in yet another dimension of complexity, cost.

How do you track the aggregate costs of each one of these services and track them over time.  How do you predict what you’ll be spending based on usage, or expected usage?  Can you do this in real-time?  And, importantly, can you take advantage of pricing differences between vendors to effect savings?

The answer to these questions as of today is “you can’t” (outside of a very few new SaaS offerings <ProductPlugComing> like uptimeCloud), it’s just too darn complicated.  Having said this, it’s important to start thinking about how to manage the emerging complexity.

 

Things to think about:

  • can you track IT services in your physical, virtual, and cloud world?
  • can you deal with these services being dynamic?  (gone are the days of a static CMDB and traditional monitoring tools)
  • can you get a holistic view of these services and their dependencies?
  • will you be able to tracks costs of these services in real-time and historically (as they are dynamic and/or across many different vendors)?
  • what hair products can you use to hide the grey hair thinking about it?

These are tough questions to answer, but are a critical starting point for adopting the Cloud.

- Alex

Tags: , , ,
Posted in cloud, cloud cost, IT Management, Systems Management | No Comments »

ShareThis
« Older Entries