We recently had time to sit down with Jai Menon, director and chief technologist for Storage Systems Architecture...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
and Design in IBM Systems Group, to discuss where storage is headed. As an IBM Fellow, Menon directs a large group of engineers and researchers at IBM Almaden Research Center in San Jose, Calif.
Menon has set an agenda of the following ten "grand challenges" in storage for IBM to tackle:
1. Managing complexity
2. Building bulletproof storage
3. Creating maintenance-free storage
4. Bringing business continuity to the masses
5. Securing storage
6. Finding or searching for information
7. Guaranteeing the authenticity of data
8. Long-term presentation of data
9. Creating an information grid
10. Replacing disk drives as the primary storage medium
While most of those challenges are self-evident (and all too familiar), a few require explanation. Number 8, for example, refers to work that IBM is doing to ensure that documents created today can be read, say, in a hundred years. That's not just a problem of reliable data storage, but of a format that can include instructions on how to read a document. That's needed because no matter how well you store a Microsoft Word document, you cannot guarantee that Word will be around in 2105 to read it with.
Information grid, number 9, is a broad concept. At heart, it's based on the notion that larger, indeed global (however you define your world) namespaces are needed, along with some mechanism for copy control. Such an information grid would allow both personal users and enterprises to define an information space within which they could establish seamless storage and retrieval of information.
Lastly, yes, IBM wants to replace disk drives. A main focal point of their work is to develop what Menon calls "storage-class memory." His goal is to create solid-state storage at a much cheaper price point than today's memory chips. One way IBM is looking at doing that is engineering chips that can store more than one bit per memory cell.
To have your say on which storage issues are most pressing at your company, take our latest poll.