RAM is the new disk
Posted 19 June 2008 @ 3:59 pm by Jim LiddleInteresting Article at InfoQ entitled “RAM is the new Disk”. The article digs into a couple of statements by Jim Gray that “Ram is the new disk” and “disk is the new tape”. It refers to an article by Tim Bray which discusses how RAM clusters are so much faster than disk clusters and his journey in the world of Grid and distributed computing for “an application” he “was thinking of”.
The article covers a broad range of technology but it would have been useful to understand the actual use case that Tim had in mind. He also touches upon the Jini Rio Framework - “It’s pretty clear that understanding Rio is going to take a lot of work, and if you’re going to be using something for infrastructure in a large-scale high-performance situation, it’s important to understand all the layers thoroughly. I think that Rio has a terrific future, but seriously needs work to reduce the barrier to entry.” Well, the GigaSpaces Service Grid is a driven dynamic monitoring and management component that was originally based upon Jini and is is pretty amazing in what it can do, and it really is easy to use as well as providing SLA driven real-time enterprise features and the power that obviously so impressed Tim.
Back to the InfoQ article, in which Nati Shalom, GigaSpaces CTO expands on the notion of RAM being the new disk explaining the limitations of database clustering and database partitioning as means to provide performance and scale saying “The fundamental problems with both database replication and database partitioning are the reliance on the performance of the file system/disk and the complexity involved in setting up database clusters.” Nati’s offered solution is the GigaSpaces In-Memory Data Grid (IMDG), backed by Hibernate 2nd level cache or GigaSpaces Spring DAO, to provide Persistence as a Service for your applications. Finally the article lists the benefits of an IMDG over a traditional DB:
- - Relies on memory which is significantly faster and more concurrent than file systems
- - Data can be accessed by reference
- - Data manipulation is performed directly on the in-memory objects
- - Reduced contention for data elements
- - Parallel aggregated queries
- - In-process local cache
- - Avoid Object-Relational Mapping (ORM)
I think the article highlights the shift that is occurring in how we think about data, where we hold it, how we reference it, and how we guarantee its availability and quality of service be that how it scales, how fast it is, or how it is distributed. All things that the folks at GigaSpaces have been talking about for a long time now….
Read more...
