Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

  1. Home
  2. Data-centers
  3. Ibm-storage-systems-research-envisions-future
  4. Page
  5. IBM Storage Systems Research Envisions the Future: Page 3 of 3
Data Centers

IBM Storage Systems Research Envisions the Future: Page 3 of 3

What is the future of storage and storage management? While all crystal balls are partly cloudy at best, one of the best bets would seem to be in IBM’s vision. Whether particular technologies make the grade or not--and, if so, when--is uncertain. However, the company’s general direction seems clear, and there are many important and exciting changes ahead. Let’s take a look at some of the developments that IBM Storage Systems Research is working on to bring about the future of s
DAVID HILL
September 14, 2011

SCM and at least several other IBM efforts deserve a separate, later discussion on their own, including these two initiatives:

  • Long Term Archive Systems: IBM is concerned about the possibility of a digital Dark Age. A 2,000-year-old Dead Sea Scroll can be read, but try to read a piece of media that is only 20 years old. The problem is divided into two parts: bit preservation and logical preservation. Bit preservation is about the ability to retrieve a bit perfect copy of digital data on an old piece of media such as a 5.5-inch floppy disk. Logical preservation is the ability to productively use the data on the bit-perfect copy that has just been retrieved. IBM is working on highly scalable bit preservation systems that can deal with billions of objects, can index for metadata and content, and is extendable for industry-specific content processing. And, yes, IBM believes that tape has an important role in the future of archiving. The company is also working on a number of other things, including automated systems to avoid format obsolescence.
  • Storage for Clouds: Of course, clouds are a favorite discussion topic in IT today. IBM’s Storage for Clouds Initiative involves a number of technical disciplines and a wide range of topics. One key focus is on geographic content distribution, where compute and storage resources can be spread across innumerable sites across the planet. Issues that are being investigated include: having data in the right place at the right time, enabling efficient access from anywhere, ensuring that any site can contribute data and ensuring data integrity across geographically distributed systems. Storage for clouds also needs to deal with such things as dynamic management of storage in virtualized environments, optimized provisioning, and real time relocation of processing and storage.

    The list goes on, such as optimizing storage for analytics and the use of shared -othing clusters in a Hadoop world. You get the idea. IBM Storage Systems Research has to
    roll up its sleeves and tackle this wide range of challenges.

    On the surface, the innovation required to bridge the gap between the explosive rise of storage requirements and the more or less fixed ability to pay for it does not viscerally convey the magnitude of what has to be done. Understand, therefore, that just evolving current storage technologies would result in bloated energy and space requirements, as well as performance that simply cannot keep up with increasing demand. All at a purchase price and management cost that would be impossibly unaffordable. Understand, too, that our concepts of private and public clouds and the Internet are very limited compared with what we will likely see in less than 10 years in terms of what IT as a service will demand in real-time access, computing developments including analytics, and information geographically spread across innumerable global sites.

    So what if IBM Storage Systems Research (and other companies, to be fair) fail to achieve their stated goals? The answer is that software and compute engines will not achieve their full potential if the storage carburetor cannot provide the right information fuel mix at the right time. We may not be able to point to the fact that something could not be done, but the ability of IT to effectively support its full measure of economic growth would be seriously compromised. So wish IBM well. Although IBM would profit greatly from a long line of successes, so would we all.

    IBM is currently a client of David Hill and the Mesabi Group.

    • Tags:

    Commentary
    Data Centers