Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure
A British company named Njini Inc. hopes to make data management easier with an approach it calls information asset management (IAM) that captures the relevance of data at creation.
"IAM is a means to interact with structured and unstructured data in real time to identify, collect and classify information such that any application can access it regardless of structure, format or location," says William Hurley, senior analyst at Enterprise Strategy Group.
Specific to Njini, the njiniEngine interrogates data at the point of origin using a process known as Identity@Origin to identify the business value of data at the moment of origin. The data is then used to understand where it should be stored, for how long, who should have access, how many copies should be kept, etc. It accomplishes this via content-aware metadata that is used to set policies.
It then drives four processes: intelligent storage that eliminates duplicates; moving data to the correct storage tier and setting how long information stays in a specific tier; compliance policing and tracking; and tagging to speed the search process. The basic idea is to give data policy so it manages itself by adding a new layer of business structure before the data is even stored.
Rather than merely being some fancy filing system, Njini seeks to turn its technology into big savings for its customers. By increasing the intelligence behind stored data, efficiencies can be realized.
"Our figures indicate typical storage costs per document run at around 30 cents," says Phil Tee, chairman and CTO of Njini. "We can cut that to about 12 cents."
The njiniEngine is supplemented by njiniEncount, a product that ensures that the correct number of each data object is retained according to its business value by replicating or de-duplicating. For example, a PowerPoint file e-mailed by one person to eight colleagues might be stored nine times in a disk array. NjiniEncount retains only one copy with eight additional pointers to that file.
"Typically, a company will have on average six duplicates of each file," says Tee. "Think of the money companies will save by de-duplicating those unnecessary files."
The njiniEngine is comprised of software that is loaded on a server appliance sitting in the data stream. It then creates a content derived database of metadata to catalog and track all data. This core engine is then supplemented by several additional modules. NjiniEncount, as covered above, handles duplicates and predefining of policies. This is the only module released to date.
In the near future, the company plans to unveil three additional modules. NjiniEnroll addresses the placement of data into the most appropriate storage tier with reference to its true business value. Classification and profiling algorithms enables migration of data from one tier to another and then another as it progresses through its lifecycle. Criteria used include the type of data, the owner, the regulations affecting it, retention periods and more.
Another module on the immediate horizon is njiniEnforce. It tracks access to data over time, ensuring only those users and applications authorized to view a file may do so. This module also enforces retention and deletion policies and leaves behind an audit trail for the purposes of compliance.
The final element of the eventual suite is njiniEnquire. This will be a way to group data so as to facilitate searches and better utilize storage resources according to business value. At the most simplistic level, this might be arranging all the Word docs or PowerPoints to be files together.
On the downside, Njini is not, for now, a standalone product. You need a decent relational database management system (RDBMS) on which to run the system. The initial version, for example, runs on Oracle 10g. Other RDBMS platforms will be supported in future. Fortunately, the company chose a database that has wide enterprise usage.
Initially, therefore, Njini will probably gain most transaction in the archiving marketplace in environments already utilizing Oracle. As the regulatory burden worsens and the demands of compliance grow, however, the allure of Njini will probably widen provided the company delivers on its product roadmap, and adds support for a greater range of platforms.
"Njini's approach will eliminate much of the need for redundant data stores and provide a more efficient means for applications to share and access data," says Hurley. "This could potentially eliminate the need for many of the existing information access and content management solutions used today."
Article courtesy of Enterprise IT Planet