Parallel Virtual Machine (PVM)

Parallel Virtual Machine (PVM) is a program that enables distributed computing among networked computers on different platforms, so that they can perform as a single, large unit for computer-intensive applications. The software can be run on most Unix-based orWindows systems. According to the official Web site, PVM is the de facto standard, world-wide, for distributed computing. PVMsource code is freely available.

PVM was created at the Oak Ridge National Laboratory in 1989 and was rewritten in 1991 at the University of Tennessee. Newer versions developed since that time offer improved portability and fault tolerance.

The PVM package has become popular for teaching computer science and programming students how distributed computing and parallel processing work. The software has been used extensively to solve problems in science, medicine, technology and engineering that have proven too complex for individual computers. PVM also enables a level of computational accuracy beyond the ability of a single computer.

physical to virtual (P2V)

Physical to virtual (P2V) is a term that refers to the migration of an operating system (OS),application programs and data from a computer's main hard disk to a virtual machine or a disk partition.

P2V can be done manually by creating or defining a virtual environment and then reinstalling the OS, applications and data on it. This can be a tedious and uncertain process, especially if the new environment contains substantially different hardware than the old. To streamline the operation, part or all of the migration can be carried out automatically by means of specialized programs known as migration tools. P2V is commonly used to accomplish server virtualization. P2V is also becoming popular as a means for Macintosh users to run Windowsapplications.

KPIT leverages VDI solution’s success

While desktop refresh cycles are imperative for high performance requirements, these cycles can prove expensive. A year and a half ago, KPIT Cummins, the technology solutions partner for manufacturing corporations faced the need to refresh and deploy its 1200 desktops across India. This resulted in KPIT Cummins evaluating various desktop technologies, which helped it zero in on VDI solutions as the best route to optimal performance and cost effectiveness.

VDI and private cloud to the rescue

With its special focus on industries like automotive, energy & utilities, industrial equipment and semiconductors, KPIT was growing considerably, and wanted to avoid purchasing individual desktops. Other key drivers behind the VDI solution selection included quick desktop provisioning, data confidentiality, regular security updates, and security benefits of centralized management. KPIT’s team realized that VDI implementation proved to be the best solution when it came to saving procurement to deployment cycle time and efforts.

As the first step, KPIT performed a thorough analysis of user usage patterns for memory, computing and other resources to reduce implementation challenges. Using capacity planning tools and open source technologies on a sample of 500 machines, KPIT collated the long term usage trends followed by in-house systems. This helped KPIT understand VDI solution requirements like the operating system, memory, disk space utilization and images to be loaded on the systems.

After assessing its requirements, KPIT evaluated several OEMs. The exercise resulted in KPIT zeroing in on a combination of VMware, Cisco and EMC’s (VCE) Vblock solutions. VMware was chosen for its virtualization capabilities, Cisco for server and networking, along with EMC for storage. “Our evaluation parameters were maturity of the technology, solution’s stability, peer references, as well as management commitment from the OEM partner,” says Sandeep Gandhi, senior program manager of global data center services at KPIT Cummins. VDA Infosolutions was the deployment partner for this VDI initiative.

VCE consortium provides system support to KPIT’s V-Block. “A single call can ensure that everything is taken care of by the individual OEMs at the backend,” says Faheem Kolhar the senior manager, client Services at KPIT Cummins. The company has signed a single license for the Vblock solution, instead of individual licenses.

KPIT has deployed HP thin clients and other devices like tablets and iPads. Windows XP and Windows 7 are the primary operating systems running on the systems.

To provide end-users with central transactions and management capabilities, KPIT built its private cloud in Pune, which hosts all major applications like SAP and Microsoft Exchange. “The decision to go for a private cloud was to ensure on the fly provisioning of servers,” says Gandhi. With private cloud deployment, KPIT claims to have reduced the administrative overhead by 80%.

The requirement for a private cloud and VDI solution was raised in Oct 2010. KPIT spent two months for a proof of concept (POC) exercise, and an additional two months to finalize the solutions for these technologies, before it went in for the final implementation.

The business stake holders for the VDI solution deployment included top management representatives like the members of senior executive management, senior delivery management, and the finance team. “VDI is a huge cultural shift for our organizations, so we wanted the consent of those directly impacted by the solution,” says Kolhar. As the ROI for a VDI solution isn’t immediate, IT management had to present a strong business case. “We highlighted the tangible management benefits post VDI deployment. This helped, as the stakeholders agreed that it will change our way of working, thereby increasing efficiency,” says Gandhi.

Cloud setup  highlights

KPIT’s V-Block includes Cisco Unified Computing Systems (UCS) with eight servers for the private cloud and 12 servers (each holding 100 virtual desktops) for its VDI solutions. VMware View is the VDI solution in place at KPIT. The end users comprise a mix of task-workers and power users whose requirements include Windows, Microsoft Office, Microsoft Exchange clients, Internet and development tools.

For the top management personnel who access applications over tablets and iPads, KPIT opted for VMware View version 5, as version 4 proved incompatible with these clients. Some of the engineering design users are not provided with virtual desktops, as they require high computing and memory power for mechanical engineering drawing applications.

KPIT performed data center power and cooling alterations to match the new infrastructure. Following the implementation, training sessions were conducted for the end-users to help better change management.

What has KPIT achieved?

KPIT has witnessed a good percent of reduction in desktop management costs, as well as power consumption costs. Instead of four administrators, the VDI solution requires just two personnel for deployment of desktops and allocation of resources. Gandhi believes that quick provisioning helps ensure instant productivity of new hires due to the reduction in time for procurement and deployment of new desktops from days to minutes. Some of the intangible benefits of KPIT’s VDI solution rollout include quick central anti-virus deployment, quick upgrades to images, and reduction in data loss risks.

KPIT estimates a five year time frame to receive direct and indirect ROI for this initiative. By March 2012, KPIT hopes to setup an identical Disaster Recovery site for the VDI setup as well as private cloud solution. “It will be located in Pune itself, but further away from the current site,” says Gandhi.

Virtual

virtual desktop infrastructure (VDI)

Virtual desktop infrastructure (VDI) is the practice of hosting a desktop operating system within a virtual machine (VM) running on a centralized server. VDI is a variation on theclient/server computing model, sometimes referred to as server-based computing (SBC). The term was coined by VMware Inc. In the past couple of years, some large organizations have turned to VDI as an alternative to the server-based computing model used by Citrix and Microsoft Terminal Services.

virtual private server (VPS) or virtual dedicated server (VDS)

A virtual private server (VPS), also called a virtual dedicated server (VDS), is a virtual serverthat appears to the user as a dedicated server but is actually installed on a computer serving multiple Web sites. A single computer can have several VPSs, each one with its ownoperating system

(OS) that runs the hosting software for a particular user.

The hosting software for each VPS can include a Web serverprogram, a File Transfer Protocol (FTP) program, a mail server program and specialized applications for activities such as e-commerce and blogging.

The VPS alternative is often chosen by small businesses that need a customized Web site but cannot afford a dedicated server. Another useful aspect of VPS technology is the ability of a single subscriber to maintain multiple virtual servers. For example a Web site owner might use one server for the production-level Web site and the other for a "dummy site" that can be used to test planned updates, modifications or new programs.

Desktop audit checklist: Five steps to a successful desktop audit

The start of a new year is an opportunity to rethink, refresh and rebuild -- even when it comes to your desktop administration strategy. Part of running a tight IT ship is conducting a regular desktop audit.

If you're like many administrators, you're always searching for new and improved ways of ensuring that enterprise desktops and client machines run free of problems at a low cost and as efficiently as possible.  It's always beneficial to reflect on the current status of your hardware and software, as well as your overall vision for management in 2012, and a desktop audit can help in these areas.

This desktop audit checklist has several ideas to get the ball rolling on your management strategy, asset management, software needs and more.

1.    Conduct a full inventory of your hardware and software assets. Even though the IT department intends to manage the entire procurement process, items are purchased -- and sometimes even deployed -- without your knowledge. (You're probably still responsible for fixing those items if they behave badly, however.) Take the beginning of this year to conduct a complete desktop audit of hardware and software. Doing so allows you to do the following:

• See exactly what you have. This is helpful for checking warranty claims, age and refresh cycles.
• Track overcommittments or underallotments in software licensing. Chances are, you've paid for too many licenses for one product, and the business may be in violation for underpurchasing on another product.
• Redeploy assets that can be used different to solve other problems -- powerful servers acting only as file shares, for example, can become virtualization hosts and color printers sitting unused in conference rooms can be shared out to entire floors.

2.    Monitor for hardware errors. Today's machines sit live, usable and turned on most of the time, and many hardware errors creep up only when machines are restarted. SMART errors -- problems with the monitoring system for SATA-based hard drives -- are easy to see upon the post-boot sequence but are more difficult to detect from within a standard Windows session. Cooling fans work until they stop spinning and then don't start again. The operating system can bypass memory errors in many cases, but become such errors are apparent when only the BIOS controls the hardware during boot. As you proceed through a desktop audit, reboot PCs, and look for these obvious errors that indicate it might be time for repair or replacement.

3.    Reimage physical Windows-based desktops on a rolling schedule. Many experts decry Windows rot -- that phenomenon where Windows systems deteriorate as applications are installed and removed and the system is generally used -- as a myth. Most industry professionals, however, agree that a clean, freshly installed copy of Windows performs better and provides less room for error. During your desktop audit, develop a schedule for reimaging client machines on a regular basis, and implement that schedule with the first batch of new machines at the beginning of this year. With a master image and simpledeployment tools, reimaging can take a matter of minutes and eliminate a massive troubleshooting burden.

4.    Banish Internet Explorer 6 and 7, and deploy IE 9.Most corporations have chosen to ignore Microsoft-alternative browses Google Chrome and Mozilla Firefox, and frankly those browser makers haven't done much to attract enterprises, with a release cadence measured in days and no coherent central management strategy. Meanwhile, IE 6 -- long the corporate standard browser -- is probably the biggest security hole in Windows, andVersion 7 isn't far behind. IE 9, the current Microsoft browser, has been out for nearly a year and is stable, compatible and secure. Moreover, its management features integrate very well with Group Policy, and it's a relatively friendly deployment. As part of your desktop audit, find out which users are still on IE 6 and 7 and make move to IE 9 on the desktop in the first quarter of 2012. (Tell your server colleagues to hold off a bit, though, as some management consoles from Exchange and other products fail on a server with IE 9 installed.)

5.    Develop a plan for, and move users to, virtualized desktops. Your desktop audit process should identify users whose daily computing needs could be met with a thin client accessing a virtualized desktop (either personalized or pooled). Call-center workers, front-line staffers with predictable workload requirements and administrative assistants generally use applications with similar performance and security profiles. With virtual desktop infrastructure (VDI), you can abstract away hardware problems and easily maintain control over users' desktop experience because a desktop lives in a single virtual machine. IT needs only to patch one machine, install software on one machine and test deployments on one machine. Make sure that a desktop audit establishes criteria for VDI suitability, and then actively develop a review process for making that transition a reality.

A complete desktop audit will help you understand your organization's current assets and take advantage of efficiencies and new technologies as they come along in 2012. Combine these suggestions with items specific to your environment as you develop your own desktop audit checklist and plans. Good luck!

Outsourcing: Glossary

Offshore outsourcing, a type of business process outsourcing (BPO), is the exporting of IT-related work from the United States and other developed countries to areas of the world where there is both political stability and lower labor costs or tax savings. Outsourcing is an arrangement in which one company provides services for another company that could also be or usually have been provided in-house. Offshore simply means "any country other than your own." The Internet and high-speed Internet connections make it possible for outsourcing to be carried out anywhere in the world, a business trend economists call globalization. In general, domestic companies interested in offshore outsourcing are not only trying to save money in order to be more price-competitive against each other, but also to enable them to compete with businesses in other countries.

Critics of offshore outsourcing worry that if too much IT-related work is farmed out to other countries, home-grown IT talent will "dry up." They point out that once a company begins outsourcing overseas, they will find it difficult to reverse the trend and justify paying more in salaries, taxes, and job benefits for the same work they used to outsource. Proponents maintain that the judicious use of offshore outsourcing will help make all IT workers become more productive and allow companies to develop more agile and responsive business models, which in turn, will raise salaries for domestic workers in all countries.

According to the IT research and analysis firm Gartner, by the year 2004 more than 40% of IT-related businesses will either be investigating the possibility of offshore outsourcing or will have already shipped some IT-related work overseas.

Supply chain management checklist for BC/DR practitioners

Most organizations deal with multiple external suppliers to support their business. One way to minimize supply chain disruptions is to learn whether your key suppliers have business continuity and disaster recovery (BC/DR) plans. More importantly, you’ll need to know how your supplier views the importance of BC/DR plans, how they manage their own plans, as well as how frequently those plans are updated and tested by the supplier.

Download this detailed supply chain management checklist, which can help you ask the right questions to assess and audit key supplier BC/DR plans and associated programs. This makes it easier to determine if your key suppliers are committed to BC/DR activities, as well as documenting and regularly testing those plans.

Evaluating supplier BC/DR plans

If a prospective or existing supplier is likely to become a key resource to your organization (e.g., suppliers of raw materials or shipping companies), be sure that the vendor has the resources and operational capabilities to stay in business even with the potential for a disruption. In both cases, examine each vendor’s BC/DR program and associated plans. The absence of such programs may not be deal-breakers, but if any of your suppliers are critical to your organization’s supply chain, their BC/DR plans could be the difference between the success and failure of your business.

You can perform the assessment yourself or use your internal audit department. The questions are designed primarily to elicit a yes or no response to minimize the amount of time needed to conduct the audit/assessment help pinpoint areas where you’ll need to dig deeper.

If at all possible, review the vendor’s BC/DR plans instead of simply taking the vendor’s word. Remember that a vendor’s refusal to let you examine the entire plan may be tied to a company security policy and should not necessarily be cause for concern. In that case, ask the vendor to provide excerpted sections from the plan, such as the table of contents and sample pages from the document.

Also be sure to assess the following when evaluating suppliers:

1. Check on the financial status of the supplier, e.g., Dun & Bradstreet reports.
2. Check on the supplier’s reputation in the community, e.g., Better Business Bureaureports.
3. Contact existing customers for references.
4. Learn if the supplier’s business has been growing, declining or staying relatively stable.
5. Review legal documents carefully, especially contracts, warranties, maintenance agreements and service level agreements.
6. Determine if the supplier has faced any legal issues, such as customer complaints, regulatory violations or litigation for poor performance.
7. Examine where the supplier’s offices, including warehouses, manufacturing facilities, and specialized facilities such as where fuel storage tanks are located.
8. Compare supplier pricing and discounting options with other suppliers.
9. Can you negotiate prices with the supplier?
10. Ask the supplier’s vendors for their opinions about the firm.

Summary

Use the supply chain management checklist to protect your organization from vendors who may not be able to maintain normal operations in the aftermath of a disruptive incident. Remember that any disruption to your supply chain can affect your organization and possibly other firms that are also in your supply chain.