Posted by Crystal Nichols on Mon, Aug 09, 2010 @ 09:37 AM
There are three pressing questions that have, and likely always will, drive data center design decisions. Accordingly, design trends for data centers can be seen as an attempt to deal with those three decisions. They include:
- How reliable must the data center be?
- How big must the data center be?
- How can the data center meet those reliability and capacity needs at the lowest cost?
In years past, many companies treated their data center like their children’s clothes. Buy it big, and they’ll grow into it. When it’s a bit tight, replace it. This led, of course, to huge data centers with unnecessarily huge power demands wasting plenty of money.
Today, data center design takes a different approach. Here are some of the rising trends in data center design:
Energy costs now equal (or even surpass) capital costs
If you can use just 10 percent more energy than what your data center needs, the average data center will reproduce its own costs five times in 20 years. Accordingly, companies are building only what they need right now, and operate at or near 100 percent capacity. This will save you in both energy costs and capital costs.
Data centers should be modular, not comprehensive
Your data center should be designed in such a way that you can add capacity without having to replace the entire data center. You should be able to treat your data center as modular, able to switch out the pieces that aren’t needed and add new ones that are. The data center should not be a comprehensive unchangeable whole that needs to be replaced.
Cooling needs take priority
The type, configuration and deployment of cooling infrastructure is more important than ever. For example, a small data center might have the cooling systems positioned close to servers, where large data centers might utilize raised flooring and chiller or cooling systems on the perimeter. Choosing efficient systems is key, as well.
Virtualization is for more than servers
Servers are usually the first component to be virtualized in the data center, but they shouldn’t be the last. Storage and other data center functions should also be virtualized in order to maximize efficiency and control costs.
Data centers are becoming smarter
Today’s data centers implement technologies such as real-time environmental monitoring, as well as equipment pre-failure notifications. The more diagnostic and troubleshooting work your data center can do on its own, the better.
No one can be sure what’s coming in data centers, but with the current pace of innovation, it’s bound to be a wild ride over the next few years.
Posted by Crystal Nichols on Fri, Aug 06, 2010 @ 09:05 AM
Virtualization can improve just about any computing environment. In a nutshell, virtualization uses a single physical server to run a number of virtual servers. Virtualization decreases your operational expenditures and helps you to utilize more of the hardware you have at your disposal. Organizations with any number of servers can benefit from virtualization. You just need to know where to begin.
The Host
The most important aspect to server virtualization when you’re in a small computing environment is the host server. This is the one physical piece of equipment that will run multiple servers. This host requires significantly less in terms of resources than multiple servers would.
Depending on the virtualization solution you choose to use, whether it’s VMware, Hyper-V or another package, you’re going to be able to run a number of servers. This is because many server types tend to run idle a large portion of the time. When a server is in use, the needs are spread out across resources including RAM, CPU, disk, network I/O and more. A virtualization server with four cores can likely run twice that number of virtual servers.
Keep in mind, of course, that some servers require a heavier workload. Database servers tend to be more intensive, for example.
CPU Needs
On average, having more cores in the server is more important than higher-speed CPUs when you’re running a virtualization server. So, a 6-core server at 2.4 GHz trumps a 4-core server at 2.93 GHz. This allows you to spread the virtualization load across more CPUs, giving you more consistent performance across the board.
RAM and Storage
After CPU considerations, RAM and Storage are next. The faster the RAM for your virtualization server, the better. It’s much harder to oversubscribe RAM than it is to oversubscribe CPU resources. Some virtual servers require a specific, fixed amount of RAM be allocated, in its entirety, to a given virtual server, so the more the better.
Storage needs can typically be met by SANs or NAS in larger environments, but in smaller environments the host server may need to handle storage. More disks are better, preferably via a RAID controller than runs RAID 5 or RAID 6.
Other Considerations
Your top concerns for virtualization are CPU, RAM and storage. Beyond that, however, you need to look at network interfaces, power supplies, the virtualization software itself and other areas of concern. Ultimately, the success or failure of your virtualization efforts may depend on how you build this first server.
The Modern Virtualizated Data Center: Click here to download the white paper!
Posted by Crystal Nichols on Tue, Jun 29, 2010 @ 10:22 AM
While there are naysayers who think that the NAS market is on the decline, there are others that forecast continued and expansive growth in the NAS market. One company suggest that the enterprise NAS market will hit 62,000 petabytes some time in 2012.
What does that mean for your enterprise? Well, for one, it means that the big NAS vendors, from EMC to HP to IBM to Dell, will all be competing for your business, alongside the traditional NAS market leaders. It also means an opportunity to take advantage of innovations in the area of file virtualization systems, and with cloud storage solutions that meet the NAS needs of your enterprise in unconventional ways.
It also means that you need to get your Enterprise NAS priorities in order. You need to recognize the problem areas in your NAS implementations. The biggest problem for the vast majority of enterprises is the rapid growth of your installed storage systems base.
You need to be able to make your NAS systems as efficient as possible in many areas, from power to cooling to footprint. How do you do that? While every enterprise has its own unique situation, there are some general guiding principles you can follow to address this enterprise NAS priority:
• Maximize your storage resource management tools. Use those tools to measure your NAS capacity and efficiency. Even starting with something basic like the File Server Resource Manager built in to Windows Server 2008 is a start. Most enterprises should consider implementing a more robust storage resource management tool, however. IBM, HP and all of the major vendors offer these kinds of tools, and there’s bound to be one that will work well in your environment.
• Focus on optimization. There are a number of optimization tools out there, as well. Specifically, you want to use storage optimization tools that will not only give you data on your environment, but be able to use deduplication and comrpession technologies, as well.
• Increase virtualization. Virtualization of your enterprise storage environment lets you use intellegent tiering of your storage, moving your less-used data to lower-cost storage systems accordingly.
• Move valuable data with infrequent access to the cloud. There are a number of solutions that allwo you to move applications to the cloud, including your storage management, optimzation and virtualization tools.
Implementing these kinds of changes will help you meet the rapidly-changing needs of your organization while making sure that your NAS solutions don’t overcrowd the resources of your data center.
Interested in learning more about how lower operational costs through process automation? Click here to download the whitepaper.
Posted by Crystal Nichols on Wed, Jun 16, 2010 @ 09:37 AM
Virtualization is one of the best ways for an organization to increase its IT service levels while reducing expenses. Whether you’re talking about server virtualization, desktop virtualization, cloud computing or server consolidation, there are many areas that virtualization can impact a business.
The challenge, for many organizations, is to be able to virtualize in an organized, comprehensive and strategic fashion. To do that, you need to be able to create a top-down virtualization strategy.
That’s not always as easy as it sounds. If technology is easy to adopt, people are more likely to get excited about it. Because of that, you need to create your strategy over time, and implement some very specific steps:
1. Address virtualization at the demand layer. Find out what the users want. Find out what the applications need. Understand where the data goes, where processing done, and go from there. Your virtualization efforts will not only be met with resistance if you don’t do this, chances are pretty good that they’ll be either ineffective or inefficient.
2. Next, you need to virtualize the supply of IT resources. This means you virtualize the network, storage, computing power, memory and IO. Each component of the IT supply has to be virtualized if a top-down strategy is going to work. Your virtual servers have to be across the board, and won’t mix well with other physical components.
3. Now, you need to incorporate the life cycle management of your virtualization platform. Essentially, you need to look at both the virtualized demand and the virtualized supply and match them with the right virtual resources. Further, you need to determine how they are matched and provisioned.
Performance, cost and efficiency are key in the virtualization process, as well. As you move through these stages, you need to always go back to those three principles. You need to look at your business processes, from most critical to least critical, in terms of those factors. From there you can decide where business needs aren’t being met, and establish priorities for moving forward.
When to Virtualize Servers
If you haven’t started using virtualization or you haven’t fully virtualized your IT environment, here are five steps to determine when you should make that move.
Click here to download white paper
Posted by Crystal Nichols on Wed, May 26, 2010 @ 08:44 AM
IBM Power Systems are some of the most valuable assets that an organization has. In order to make the most of those assets, however, you need to be able to understand how the system is being utilized, and how your organization’s growth is likely to impact that system. You need to be able to make realistic plans on how to improve service and manage risks, all while reducing costs. Power Systems provide you with a framework in which to meet the changing needs of your organization, and IBM’s Performance Management for Power Systems gives you the ability to do just that.
Here are some ways that you can use PM for Power Systems to maximize your organization’s configuration:
Visualize your Virtualization
PM for Power Systems lets you step back and look at the big picture. You can see what utilization truly is across the enterprise. From there, you can make intelligent and informed decisions about virtualization. You can decide, for example, whether to implement a multi-OS multi-partition environment. PM for Power Systems gives you the information you need to make that kind of decision. You can even use PM for Power Systems to try out various system workloads on your IBM Power System.
Ongoing interactive Access
Not only can you take a snapshot of what your Power System utilization looks like right now, you can also have access to historical performance data. This means that you can look back at utilization and capacity concerns as far back as 24 months, which can be extremely useful in making plans for future utilization.
Easy Configuration
PM for Power Systems comes in a couple of different options. There are “no additional charge” and “nominal charge” options, depending on the level of detail that you need from your systems.
Once you complete the initial setup process, the rest of what PM for Power Systems does for you is automated. This relieves valuable IT personnel from the tedious, laborious and, ultimately, expensive tasks that make up the day-to-day systems management routine.
Broad Support
If you have a mix of different IBM technologies, PM for Power Systems can provide you with integrated access to all of your hardware configurations. Currently, PM for Power Systems supports current IBM I as well as AIX and the hardware models that they are currently running on. It also supports IBM POWER processor-based blades running both IBM I and AIX. It doesn’t support x86 blades, and it doesn’t support a handful of rack type servers. Make sure yours are supported, of course, before you implement PM for Power Systems.
5 steps to determine when to Virtualize.
If you haven’t started using virtualization or you haven’t fully virtualized your IT environment, here are five steps to determine when you should make that move.
Click here to read more!
Posted by Crystal Nichols on Mon, May 10, 2010 @ 06:26 AM
The sheer number of choices available to today’s enterprise can be mind-boggling. In spite of that fact, and depending on the size and the particular needs of your organization, you might have a hard time wading through the various options to find the one that really fits your needs.
For many enterprises, IBM Power Systems are a good fit. Power Systems can, for some, offer a lower TCO and provide you with an infrastructure that’s more resilient than most other configurations.
Simplicity
One of the things you’ll find with Power Systems is a simplicity not often found in today’s complex world. The IBM technology roadmap is innovative, dependable and scalable. In addition, IBM technologies allow you the simplicity of scaling your systems up via upgrades or scaling your systems within using PowerVM and Capacity on Demand technologies. IBM also offers migration expertise for enterprises moving from other platforms, including x86, SUN and HP.
What this all means for your organization is that you don’t need to allocate more resources than necessary to meet your IT needs.
Maximized ROI
The design of the core technologies that the Power Systems roadmap is built on means that you aren’t going to pay for more capacity than what you need. IBM’s virtualization technologies will reduce your costs and help with data center consolidation. You can virtualize everything or nearly everything in your data center, meaning you never buy more hardware than you need.
This also means that you get consistent performance for your money. Lost productivity due to planned or unplanned outages can become a thing of the past, which will give immediate and measurable ROI.
Optimized Value
Power Systems’ reliability, serviceability and availability mean that you can not only avoid unplanned downtime, but minimize planned downtime. A trio of technology options let you optimize your operations: Capacity on Demand, Hot-Node Add and Hot-Memory Add. These technologies let you maintain availability for your most important applications while you add capacity.
In addition, Power Systems have been optimized to run multiple applications in a secure environment across OS’s on a single server. This means that you can have i, AIX and Linux all running on the same server. This means there are fewer systems to manage, lower capital costs and higher resource utilizations. Those complex and inefficient server farms are going to be a thing of the past.
IBM GLOBAL CIO STUDY: DOWNLOAD TODAY!
After thousands of interviews, we found that successful CIOs actually blend three pairs of roles. These dual roles seem contradictory, but they are actually complementary. To characterize each role, we have coined a term that describes its dominant quality. At any given time, a CIO is:
• An Insightful Visionary and an Able Pragmatist
• A Savvy Value Creator and a Relentless Cost Cutter
• A Collaborative Business Leader and an Inspiring IT Manager
Read more by downloading the white paper.
Posted by Crystal Nichols on Fri, May 07, 2010 @ 08:36 AM
Cloud computing is one of the best ways to reduce your IT cost and complexity all while helping to optimize your workload and provide the highest possible availability to your user base. Cloud computing utilizes a dynamic infrastructure that is specifically designed to provide more services and capacity while using fewer server resources. One of the best ways to implement a cloud computing solution is with IBM Power Systems.
In fact, IBM Power Systems is in an ideal solution for cloud environments. There are a number of ways in which IBM Power Systems can be the foundation for your cloud, and a number of options to get you there.
The Right Kind of Workload Optimization
At the core of cloud computing is this idea of optimizing the workload. This allows you to make the most of your IT resources while increasing your overall flexibility. Power Systems use technology like IBM’s New Intelligent Threads to switch between processor threading dynamically. The Power Systems TurboCore mode lets you provide the most performance per core for things like database or transaction workloads. Active Memory expansion lets you expand your physical memory logically by as much as 100 percent for memory-intensive workloads like SAP.
Limitless Virtualization
With PowerVM, the virtualization component to IBM Power Systems, you can virtualize not just processor resources, but memory and I/O resources as well. You can use PowerVM to adjust capacity in a dynamic fashion, to move workloads between servers, and to maximize availability. This kind of virtualization even allows you to prevent planned downtime.
Automated Management
Being able to provision resources within the cloud is key to maximizing utilization and efficiency. It also helps to reduce your TCO and management costs. Utilizing IBM Systems Director Enterprise for Power Systems, you have a way to manage physical as well as virtual servers in an automated fashion. These tools are cross-platform, too. This means that, no matter what your environment, the Power Systems cloud can provision virtual machine images and effectively allocate resources, all while providing you with an accurate picture of how your systems are operating.
Solutions of All Kinds
No matter the shape, size or composition of your cloud, IBM Power Systems has a possible solution. Here are a few of the specific offerings:
• IBM CloudBurst. CloudBurst lets the data center quickly create and implement a private cloud environment. It’s a cloud computing quickstart aimed at a defined portion of the data center.
• IBM WebSphere CloudBurst Appliance. This offering lets you deploy and manage your SOA foundation in a cloud computing environment, and easily deploys WebSphere virtual images to your Power Systems partitions.
• IBM Smart Business Development and Test Cloud. This solution lets you create a private cloud environment for the purposes of development and testing, reducing your operating costs and your test cycle times.
Download the IBM Cloud Computing White Paper
In recent year
s, cloud computing environments have been gaining popularity.
To better understand the current rate of adoption, as well as drivers, barriers and considerations that are influencing the adoption of cloud computing, IBM conducted a survey in June and July of 2009 of 1,090 IT and line-of-business (LOB) decision makers around the world. Download the white paper to see the results! Click here for more information
Posted by Crystal Nichols on Mon, Apr 26, 2010 @ 08:31 AM
Virtualization isn’t just a hot trend. Virtualization is at the center of solid IT strategies across many industries and around the globe. Virtualization lets you take all of the workloads across the enterprise and put them onto fewer systems. This, in turn, reduces your capital costs and increases your server efficiency and utilization.
In the world of virtualization, IBM PowerVM is one of the top contenders. If you want to put the power of virtualization to work in your organization, you should be looking at this cutting-edge technology.
Here are 3 ways that IBM’s PowerVM will benefit your business:
1. Utilization
Increased utilization of resources. PowerVM lets you make the most of your data center resources. Not only does PowerVM maximize your utilization, it also does it in such a way that is more effective and efficient than other VM technologies. With virtualization machines from other vendors, each physical core has its own queue. This means that each processor has to do its own work, and that requests are sent into queues based on workload. If one core finishes its queue, you may still have other processes waiting in other queues.
PowerVM utilizes just a single queue. When a core opens up, the request that's at the front of the line goes to the next available processor. In this way, IBM PowerVM makes the most of your hardware and reduces the possibility that you'll have physical cores sitting idle while requests go unanswered in a different queue.
PowerVM also works with the physical hardware in a Power System to make the most of memory usage. Two IBM technologies, Active Memory Sharing and Active Memory Expansion, help with this. Active Memory Sharing works in a manner similar to the core queue system in power VM. Active Memory Expansion utilizes memory compression and decompression methods to make it seem like physical memory is larger than it is within a given memory partition.
2. Availability
One of the concerns many enterprises have when implementing a VM technology is availability. PowerVM increases the number of workloads on a server, which raises the concern that a process could, potentially, be disrupted. Not only that, when a server goes down there is the potential that more workloads and more users will be affected.
The good news here is that powerVM and IBM Power Systems have a number of features aimed at keeping availability at a maximum. PowerVM provides redundancy through a number of technologies, including shared I/O and multiple VIOS partitions. This eliminates single points of failure in your I/O subsystem, meaning that hardware upgrades, repairs and reconfigurations don't need a planned outage. You can move an application rather than stopping it.
3. Flexibility
PowerVM can handle servers on a broad range of platforms, from Linux to UNIX to Windows. Of course, it supports the full range of IBM servers, including the new POWER7 systems. The competition uses specific VMs for specific server lines. That means that if you reconfigure your data center by getting rid of smaller servers and moving to larger servers, you need to reconfigure your partitions. With PowerVM, you don't have that kind of limitation.
Posted by Crystal Nichols on Fri, Mar 19, 2010 @ 08:30 AM
Storage management, like so many other areas of technology, is changing on an almost daily basis. Those involved in storage management, from the end-user support specialist up through the CIO, need to have a handle on what’s going on so that they can provide the best possible results to their business.
Here are 5 of the most significant storage management trends you should be familiar with:
Life Cycle Management
Information life cycle management is the process of actively managing your data from the time that it’s created all the way through the process until it’s not ever needed. Life cycle management is one of the most effective ways to reduce your overall storage costs while making better use of your resources. You can classify your company’s data according to how business critical it is and then put that data on the most reliable (and most expensive) storage. Less important data can then be kept on less expensive storage. Life cycle management lets you create multiple tiers of priority, and ultimately reduce infrastructure costs.
Consolidation/Virtualization
For the past few years, virtualization has been at the center of just about any discussion of data management, and that trend continues. By reducing the number of servers, as well as the kind of servers you have on site will help reduce the complexities involved in storage management. Virtualization also lets you place data in locations that are most conducive to the critical performance needs of the enterprise.
Capacity Planning
Constant monitoring of your storage utilization is key to being able to predict and plan what your storage needs will be in the future. Capacity planning involves more than just volume size; it includes things like bandwidth and CPU utilization, as well.
Centralization
Here’s another trend that’s been hot for several years, and that continues today. By having a centralized team manage your storage requirements, you wind up with fewer overall hassles as well as the ability to aggregate your available capacity into usable and managed pools.
Resource Management
SRM is finally catching up with us. SRM is more than just tracking and monitoring backup data, however. It’s all about taking all of your data-related needs and managing them from the time the data is generated through the entire lifecycle.
Manage More Data with Less Infrastructure
Click here to download the white paper: Using IBM data reduction solutions to manage more data with less infrastructure
This paper addresses the challenge of managing and surviving the tidalwave of data and describes various options available from IBM to deal with the challenge by effectively reducing the amount of data that must be managed.
Posted by Crystal Nichols on Fri, Mar 05, 2010 @ 06:18 AM
Virtualization is one of the hottest trends in IT, and for good reason. Virtualization can help save your organization plenty of money, as well as reduce your overall ongoing staffing, support and maintenance needs.
Still, virtualization isn't a universal remedy. There are several things virtualization can't do for your organization, and understanding the limitations of virtualization is as important as understanding the benefits.
Here are some of the biggest virtualization myths:
1. Virtual machines have higher availablility and reliability. This is one of the "features" of virtualization that vendors often talk about. They point out how applications can easily be restarted during an outage, for example. Unfortunately, virtualization also creates a single point of failure in the one physical machine. Complement your virtualization with orchestration management, however, and you can actually wind up increasing both availability and reliability.
2. Virtualization will allow applications to perform exactly as they did before. The fact of the matter is that performance is likely to degrade. The degree to which it degrades depends on the physical hardware of the server, as well as the rest of the server environment. In many cases, the degredation will be so small as to be irrelevant, however.
3. Virtualization always saves money. To be sure, most virtualization implementations will reduce the total cost of ownership. However, it's important to recognize that some of the potential savings have to do with staffing levels. If you virtualize a great deal of your environment but still wind up having to staff the same number of admins or engineers, or if you're not willing to downsize some, you're going to wind up paying more in the long run.
4. Virtualization is important for SOA. Some SOA applications do provide good prospects for virtual machines. The application services can be encapsulated, and then hosted on as many machines as the organzation needs. Still, there are many other factors as to whether virtualization will perform for SOA, including communications architecture, application design and host configuration.
5. Virtualization is more secure. In some ways, virtualization makes security resposne more effecient. If a single virtual machine is breached, then that machine can be stopped and reloaded without impacting other virtual machines. Still, the virtual machine technological layer may or may not be secure in and of itself, and virtual machines that have been compromised could have also corrupted shared data.
6. Any organization can be ready for virtualization. There are many issues that have to be addressed before virtualization can be considered, such as managment team issues, development issues and process issues. This is to say nothing of the potential human impact of virtualization, and the political fallout it could cause in a larger environment. A rush to virtualize can cause catastrophic damage in other areas.
7. Virtualization doesn't require specialized knowledge. Regardless of what proponents might say, adding an entirely new layer of software and migrating applicaitons from individual servers to a single larger machine takes some knowledge, both in terms of the virtualization technology and in the application area.
