Disaster Recovery

Classification + protection = a safe organisation

As 3DR makes its way into the Indian market, technologies like CDP and ILM seem to be the tools for organisations aiming to make sure that their data is classified well and protected better. Rishiraj Verma reports

It’s an organisational imperative to make sure that data is available at all times. Another important objective is to ensure that the available data is also secure. It is to help organisations accomplish these two goals that technologies such as CDP (Continuous Data Protection) and ILM (Information Lifecycle Management) enter the picture.

Here we take a look at the roles of CDP and ILM in the enterprise when it comes to data classification and the ability to continually protect it.

CDP: Concept and Functioning

The data that organisations store today is rising at very high rates. This increasing quantity and the always-on nature of organisations make them realise the need to protect data constantly.

Be it an insurance company with its list of claims or a BPO firm with its numerous clients, the aim is to keep all this increasing data safe at all times. Another motivation to protect data is compliance measures, which may call for any bit of data at any point of time.

CDP is one of the most helpful concepts currently in vogue. Manufacturers insist that CDP is a concept and not a technology, which is what makes defining the term difficult. “However, in its purest form, the concept means the ability to restore data to about any point in time. This may be done by applying the concept to the organisation’s replication software, or as a stand-alone software,” says Heidi Biggar, Analyst at the Enterprise Strategy Group.

It’s different. The basic function of CDP is to make sure that all modifications in an organisation’s data are recorded at the time of change, as stated by Biggar. This granular manner of back-up is meant to ensure that if a file or entire system is affected by a virus, trojan or even disaster, the most recent copies can be retrieved.

The term ‘continuous’ therefore is what marks out CDP from other data protection technologies available in the market today. As Soumitra Agarwal, Marketing Director, NetApp India explains, “CDP can give you more frequent back-ups and rapid restores.” This, according to him, can drastically shrink the back-up window and save the organisation much valuable time.

Says Avijit Basu, Country Marketing Manager, Enterprise Server and Storage, HP, “The biggest difference with CDP is that it is application-centric.” He states that CDP basically “adds a lot of speed to back-up and restore processes.”

Rajendra Dhavale

There are other major advantages which CDP can bring to an organisation. Rajendra Dhavale, Consulting Director with CA, points out the incremental nature of back-ups that CDP facilitates. “You can roll back even a few minutes back in time to restore modified data.” Biggar further states that CDP is similar to snapshots in many ways, but it is more space efficient.

These comments may be able prove that CDP, while protecting data to what may be termed an optimum level, also acts as a time-saving factor for organisations. This dual advantage is probably the reason why a large number of organisations are moving towards CDP.

Challenges it may solve. It is a known fact that protecting data can be a challenging task, not just for large enterprises but for small and mid-sized firms as well. Market vendors told us of some of the problems they face on the path to protecting their data.

Vivekanand Venugopal

Vivekanand Venugopal, Director, Software Solutions, APAC, Hitachi Data Systems, specifically spoke about the SMB segment. “These organisations work in a manner so that they cannot be dependent on just the primary site for back-ups.” He says that most of the organisations that fall in the SMB segment are bandwidth-sensitive. This, according to him, may cause them to look at only a limited number of options to regularly back-up their data.

Agarwal points out that security may be a major challenge that organisations face while protecting or even while backing up data. “It is highly possible that an insider threat may occur during the process of back-up.” He says that critical data may fall into the hands of unauthorised employees, and hence the need for constant backing up.

But as Basu remarks, “Back-up challenges are different for different types of organisations.” According to him, larger organisations may be more concerned about their large databases and want to cut down on the number of back-up windows. SMBs, on the other hand, look for flexibility and reliability of the back-up applications they use.

Sunny John

Sunny John, Country Manager for India, Quantum, has something to add. “With regulatory compliance requirements in place, many businesses today are legally required to use formal data protection and thus are becoming increasingly concerned about protecting their data.” He points out that losing key data may result in huge losses to an organisation.

A number of organisations use data that needs to be updated regularly. It is such data that needs to be protected in an even superior manner. Talking about data loss, while important information is being worked upon, relying only on a nightly back-up system may prove fatal to the organisation irrespective of its size. Thus, another detail goes on to prove the growing importance of CDP in the enterprise.

Future CDP. With positive remarks coming from analysts and vendors, it can be said that the future of CDP definitely seems bright. Organisations have already realised the importance of their data and protecting it. Soon they may realise the importance of protecting it on a continual basis rather than a daily or weekly one. This is one way of making sure that CDP doesn’t get written off as ‘just another buzzword.’

ILM today

Talking strictly from the DR perspective, ILM is a term that has been much talked about in the industry for a couple of years now. The concept’s major aim is to make sure that organisations achieve the most cost-effective means to store the most important data.

Of the many functions that ILM as a concept can provide to the organisation (vis-à-vis automation, review, classification, etc), the most important one, according to manufacturers, is classification. Vendors say that it is this aspect of ILM which makes sure that the right data is backed up to the right level and stored for the right time.

Basu comments, “From its birth to death data goes through various stages, and ILM plays an important role in giving character to this data, marking it important, critical and so on.” He explains that ILM is slowly moving on from being just data classification to being more specific. “ILM’s focus is now specifically on database archival.”

But some people are also of the view that ILM is a lot of hype and not much is really happening on the front. Insists Venugopal, “The concept is pretty hyped. There aren’t any full fledged implementations around.” He feels that the reason for this is the lack of data classification tools.

Agarwal is a little more optimistic. “Till very recently it was all talk, but now a few customers have started to look at it seriously.” He explains that ILM may be very helpful to the organisation as far as DR is concerned, but technologies available as of today may not be sufficient to make the concept work as a whole.

John concurs. “The most critical and difficult part of ILM is the classification of information according to its importance, but solutions to perform this critical function are not mature yet.”

Looking at these views, it can be said that ILM may have achieved the status of ‘most-talked-about-concept’ in the industry, but it still has a long way to go as far as its real implementations are concerned. What may be needed is the right set of technologies to prove that the concept can really hold water when supported with the correct technology.

The advantages of ILM. There are however positives which the concept can bring to the organisation. According to Agarwal, “ILM may be able to assure the organisation of data integrity and data authenticity.” He adds that this concept may be able to make sure that regular back-ups are taken without too much crammed information.

Venugopal talks about a fluid storage structure. “Application classification and optimised storage of applications is what will give organisations a good storage structure.” He links these points to ILM, saying that the classification function which ILM brings along with itself can help organisations here.

Dhavale simply says that an organisation with ILM versus one without it could be as good as structured processes versus random tools and technology. According to him, this is the mindset that organisations need to acquire if they have to set the data priorities right.

The larger picture. Like any other ‘good’ concept, ILM too may not be able to work in isolation. Many believe that ILM works as a complementing concept for others such as 3DR. This could make organisations think of ILM as a necessity in the near future rather than as just a concept that they ‘could’ implement.

So how can ILM fit into a 3DR set-up? Will it serve its purpose if it works with 3DR? Are there any hindrances in adopting the concept? Vendors try to answer these questions.

Opines Agarwal, “The idea of a 3DR set-up is to make sure that data—and business critical data at that—is always available and secure. ILM may be helpful here in that you can use it to understand what data is more important than others and needs to go on tape.” He says that this will act as a major cost and time saving factor for organisations which have a lot of data to take care of.

Dhavale feels that any organisation today needs to have process-oriented management of its data since “there aren’t any specific standards as far as lifecycles of information are concerned.” He says that this is the reason why organisations need to adopt this concept so that standards can be set and all organisations are able to manage data carefully. As far as 3DR is concerned, he is of the opinion that it is 3DR that fits into the ILM set-up and not the other way around. “Your 1D, 2D and 3D are all decided when you classify and index data according to ILM.”

John points out that “ILM fits well into a 3DR set-up with each class of data residing in the right type of storage with the right back-up and recovery strategy.” He feels that ILM helps in deciding the most critical data and therefore everything else can be dumped on tapes, thus saving costs.

Basu feels that ILM can definitely fit into the 3DR set-up but “the ILM-3DR marriage may take a little while.”

What may happen. As of now, the future of ILM may be uncertain, given the fact that most vendors are of the belief that it is only talk. A possible connection with DR however is a different story altogether.

Agarwal remains optimistic and says that a lot of third-party solutions and software may help ILM on its way to becoming an integral part of 3DR. “I feel that the BPO and IT-related industries will be the first to adopt this concept. They always need to reduce costs.”

Basu feels that ILM has definitely improved over the past few years. According to him, the search engines used for retrieving data have now become much stronger. He thinks that the future is positive for ILM. “Much software is getting evolved, particularly to help ILM. Many vendors are also looking at e-mail and database archiving. There is also more focus on the DR perspective.”

On the money factor John says, “Because of the decreasing price of low-cost disc systems and virtual tape libraries, they may be used as secondary storage according to ILM and for fast restores in 3DR.”

So while scepticism follows ILM to its deepest roots, there may still be hope for its future.