RSS FeedWhite Papers

White Paper Download

Replication as a Service

It's a whole new lease of life

Category: Virtualisation

Date: , 15:05

Company: SunGard Availability Services

The security of having a single version of data, mirrored on one or more servers has been a privilege only the biggest corporations could afford. But the launch of SunGard’s Replication as a Service makes this vital technology available to a whole new customer base.

Executive BriefingReplication as a ServiceIt s a Whole New Lease of Lifebusiness technology leadershipSPONSORED BYUntitled DocumentThe security of having a single version of data, mirrored on one or more servers has been a privilege only the biggest corporations could afford. But the launch of SunGard s Replication as a Service makes this vital technology available to a whole new customer base.Replication is pretty well understood in any fast moving industry. In a sector such as financial services, where banks trade billions of dollars every second, a mere blip in trading is not an option. Downtime is unthinkable. System crashes could literally spark a chain of events that would mortally wound the western economy. So a period of downtime, followed by a delay of two days, while the banking system is restored from its last backup, has not been acceptable for most global corporations for decades. Replication of data protected them from this danger.But global corporations could afford to invest in exact replicas of their production environment, server for server. They would spend whatever it took to create instant replication of data across all parts of an organisation. Replication means that no matter how many transactions are taking place at any one time, there is more than one version of this information stored on a different data source, updated to within a millisecond s accuracy. Rapid recovery must be guaranteed from any event. This is achieved by having multiple data centres, all driven by state of the art storage systems and processing engines, linked by the fastest communication links available.With the launch of a managed replication service, SunGard is signalling the start of a new age in replication.Prior to the launch of SunGard s Replication as a Service the cost of replication had been prohibitive to CIOs because of the expensive duplication of an organisation s live environment. Adding to the complexity of replication requirements and the management of data replication is the incompatibility of the systems that companies will have used as their infrastructure grew. Many companies find themselves with a mixture of storage media. All of which are proprietary and unable to communicate with each other.With the launch of Replication as a Service, the price barrier of this vital business requirement and technology has been removed. The economies of SunGard s service are now being passed on to a new range of corporate customer.The purchasing of hardware and software are huge capital expenses to a CIO. A managed service represents a budget outgoing and can be incorporated into the CIO s operational expense, which is currently a popular train of thought with CIOs and CEOs for good business reasons. But it s not just the hardware costs (which can be hire purchased) that you are able to OpEx, the people resource and real estate costs can also be better managed and often less.Enterprises have required replication for sometime, but they needed a more cost effective means of achieving it. The current economic logic of the CIO world suggests they should outsource this activity to a specialist who can achieve it more efficiently. The launch of SunGard s Replication as a Service is a sign that the market is reaching a new stage of maturity.For organisations new to replication, a service provider makes obvious financial sense. Indeed, for companies that already have their own replication systems, handing over to a service provider has a strong business logic too. A service provider frees the buyer from the needless duplication of manpower, investment and management. Better still, being specialists, they have the best skills and experience available. Whatever challenge faces a company their service provider has been there, solved it and perfected the technique. A failed routine in a replication process is a potential disaster unless somehow it can be averted. Normally this is a slow process while the source of the problem is investigated. The search for the rogue IP addresses or protocols which the system does not understand can be agonisingly slow and painstaking when you are under pressure. To avoid this SunGard s Replication as a Service enables the asynchronous replication of data from multiple vendors to a single resilient storage environment. SunGard s Replication as a Service technology also homogenises data so that the platform of its origin is not an issue, therefore offering the CIO peace of mind. At a conservative estimate, a company can slash up to 38 per cent of the cost of its replication strategy by outsourcing it to SunGard. The study (conducted by SunGard) compared the costs of managing a standardised storage platform with outsourcing. In mixed environments, the savings could be even higher. 2Untitled DocumentReplication and the business case for Replication as a Service In day to day operations, replication will return your investment. In a disaster, it will play a crucial role in the connected recovery of your business. Without it, enforced downtime and loss of critical data could be fatal. Any company that already stores its data might assume that replication is a luxury investment. They have the data, so they assume that restoring the systems will be a formality. The verdict, from veterans of the process, is that it s a lot more complicated. Ideally, this is a service that is better performed by experts.With analyst IDC reporting that the average company s data grows by 45 per cent every year, managing data is becoming increasingly time consuming and expensive. This is one of the by-products of the web s influence on business, with so much more commerce being conducted online. Another of its legacies is the change in legal compliance. The pace of change forced by the Internet has caught legislators napping. To compensate, they are now bombarding companies with regulations and constantly evolving ISO standards for data protection strategies. The responsibility for which rests with the CIO. This added pressure couldn t have come at a worse time for the CIO, when budgets have never been under so much pressure.The replication of data gives you two or more copies of all the data in your company. Every stage of every transaction and every piece of static information is mirrored or copied onto at least one other storage system. Co-ordinating the collection and storage of this data is subsequently a massive task. It s been made far more complicated by the distributed nature of data in many companies. A less kind description might be all over the place . The growth pattern for most company storage infrastructures could best be described as organic. As data increases every year, the servers that contain them have multiplied like Japanese Knotweed, although Knotweed at least comes in one species. Storage hardware comes in seemingly infinite varieties, none of which are compatible with the others. Trying to marshal all of a company s servers into one cohesive unit is a challenge, especially as HP, EMC and IBM units (to name but a few) have mutually exclusive languages and probably always will do.Without these complications storage still demands a considerable investment in hardware and software, not to mention the specialised staff needed to manage and operate the system and the real estate to house the replicated environment. The proprietary nature of the systems multiplies the effort needed to create a coherent management system that can collate and store all this data every second of the day.The expense is entirely justified, however. Any company that fails to replicate data can pay the ultimate price going out of business if their IT systems crash. In certain industries, such as financial services, the people held responsible for the loss of value (to shareholders, customers and stakeholders) could be held criminally responsible. There are also positive reasons for having a replication strategy. Compliance with standards and regulations isn t just about saving money on fines. It creates the framework for a well ordered system that brings its own efficiency savings. Replication is about safe storage and retrieval of information. Storage, management and retrieval of information are the key skills vital to the information age. Logically then a good replication strategy is a good foundation for competing in the digital economy.A replication system and associated recovery mechanisms are measured in terms of recovery time objectives (RTO) and recovery point objectives (RPO). As with all trading systems, they have values set according to the priorities of each business operation. Each is determined by a set of criteria, such as: How up to date is the data when it s replicated, where the data is stored and how quickly can it be restored?The ultimate question, in calculating commercial risk, is this: How much data loss would be acceptable to my business? In some cases it is a minute, in some a day and in some it can be no more than seconds. The business case 3Untitled Documentfor replication is about making judgements about every aspect of data at every aspect of the production process. These judgements themselves are a useful audit of how the business actually functions.By being forced to carry out a data replication process, a business can be forced to make an assessment of the very fundamentals of its commercial practise. An audit of data becomes an audit of the business and a health check on its functions. In industries where downtime and lost data costs millions of pounds, the regulators have caught up and created rigid standards of compliance that must be adhered to. In 2006, a global investment bank was fined 15 million by regulator the Securities and Exchange Commission after it failed to retain tens of thousands of emails sought in cases against the brokerage firm. Saving 15 million would be a reasonable return on replication investment.To make the case for Replication as a Service even more persuasive, the process has been refined to keep costs down. Not every item of data has to be replicated entirely. Much of the data in any system is not indispensable. Some data contains duplicate records, some contains outdated records. The process of replication updates all of an organisation s required data. Data that has already been replicated is not over-written, but of course remains ready for retrieval. Until recently, enterprise-standard replication has not been affordable to the majority of businesses. With the invention of SunGard s Replication as a Service, this function could be within their budget. But CIOs will always be pressed to prove that there is a business case for replication. Here is a simple example of the business case for a medium sized enterprise.A company with 250 employees with a storage area network (SAN) in place and a tape back up routine may assume this is sufficient protection. But there are fundamental flaws in this plan. The first business cost is the risk. With all data on site the business is prey to fire, floods and other disasters or simple denial of access or hardware failure. Another cost is the inevitable mixture of incompatible and uncommunicative storage devices, from competing vendors, that will hold the data. As described above, any short term savings achieved by shopping around are neutralised several times over by the management headache created by a clash of proprietary systems. Each vendor s replication system recognises only its own kit. So the cost of the overall process of management is multiplied, based on purchase of multiple systems, that are rarely at maximum use, and a labour intensive process.As if that were not enough, there will be a third management cost, albeit one that frequently doesn t appear on spreadsheets. The man hours spent on tape back ups are phenomenal. Tape has another cost, in that this outdated technology is the basis for a lengthy restoration process. Given the pace of business these days, few companies can survive two days of downtime while the systems are restored from a tape back up. That s assuming that a system can actually be resurrected using this method. It s rarely tested, as it takes too long. A system crash is not the best time to discover that tape restorations don t actually work because the tape has degraded.In the event that a disaster happened between backup cycles it is inevitable important data will be lost. According to analysts at IDC, the majority of SMEs never recover from a 24 hour period of downtime and go out of business within a year. SunGard s Replication as a Service means, that whatever size your company, it can come back to life in seconds.The SunGard Replication as a Service not only delivers a return on your investment. It s about getting the return of your business. IDG 2010. IDG Communications Ltd, 101 Euston Road, London NW1 2RAAbout SunGard Availability ServicesSunGard Availability Services mission is to help customers prepare for and recover from emergencies by helping them minimalise their computer downtime and optimise their uptime, despite situations that threaten to interrupt their business.More than 10,000 customers worldwide rely on SunGard Availability Services to achieve uninterrupted access to the information systems they need to do business. Trust us to do the same for you.Tel: 0800 143 413Email: infoavail@sungard.comWeb: www.sungard.co.uk

You must have an account to access this white paper. Please register below. If you already have an account, please login.

Already registered?

Login

Forgot password?

New customer?

White paper download

ComputerworldUK Webcast

ComputerworldUK
Share
x
Open
* *