Coherent, Gigaspaces or Persistence?

Discussions

Performance and scalability: Coherent, Gigaspaces or Persistence?

  1. Coherent, Gigaspaces or Persistence? (3 messages)

    I am looking into those vendors for a distributed caching solution.

    They do provide similar caching functionality but their underlying technology seems not that the same.

    Our system already have a good O-R mapping layer and looking highly available caching solution for performance and scalability. We may also need some integration with intelligence load-balancing as some of the cache object maybe very large and we want request be routed based on the location of the cache as well as other factors.

    Anyone here have done related study before? Your information is very appreciated! Thanks!!
  2. Coherent, Gigaspaces or Persistence?[ Go to top ]

    Mr. Hong,

    There is a big difference between the products you mentioned and GigaSpaces.

    The GigaSpaces grid server is targeted specifically at real-time distributed applications requiring both fast data access to distributed information, and the ability to process this information in a timely fashion - generally via parallel processing.

    The GigaSpaces grid server addresses the challenge of processing in a grid architecture - i.e. how to provide scalability, redundancy and performance by virtualizing memory resources.

    The GigaSpaces grid server provides tightly integrated messaging and caching capabilities - ensuring consistency, efficiency, and dynamic scalability, as well as runtime management capabilities.

    GigaSpaces is the only grid-based solution providing such tight integration between messaging, caching and pure shared memory capabilities.

    GigaSpaces caching supports both the Sun JavaSpaces standard as well as the Java HashTable API - which is considered the de-facto standard for cache API. GigaSpaces provides a JDBC/SQL adaptor, enabling sophisticated data manipulation and a full JMS implementation on top of its data grid for messaging support.

    See the following links providing some insight into GigaSpaces Core capabilities:
    http://www.gigaspaces.com/docs/doc/index.htm#gigaspaces_cluster.htm
    http://www.gigaspaces.com/docs/doc/gigaspaces_distributed_caching.htm
    http://www.gigaspaces.com/docs/doc/index.htm#hibernate_cache.htm

    None of the products mentioned provide such solutions for your requirements.
    We will be happy to setup a conf call with you to provide you a full comparison between the products you mentioned and GigaSpaces.
     
    Best Regards,
            Shay
    ----------------------------------------------------
    Shay Hassidim
    Product Manager, GigaSpaces Technologies
    Email: shay at gigaspaces dot com
    Website: http://www.gigaspaces.com
  3. Coherence, Gigaspaces or Persistence?[ Go to top ]

    Hi Feng,

    If it is customer input that you are looking for rather than marketing information, you can take a look at our newly published customer stories. These are the first two and there are many more in the pipeline.

    Later,
    Rob Misek
    Tangosol, Inc.
    Coherence: It just works.
  4. Hi Feng,
    I am looking into those vendors for a distributed caching solution.
    A free evaluation copy of Tangosol Coherence can be downloaded from http://www.tangosol.com/download.jsp.
    They do provide similar caching functionality but their underlying technology seems not that the same.
    It is true that all three products discuss caching as a feature, but they are all very different. Tangosol Coherence has been a pure Java clustered caching product since its inception, and Tangosol is actively involved (as the spec lead for JSR 107) in the standardization of caching in Java and J2EE. Persistence has an integrated ORM + caching solution with tie-ins to model-driven development tools, and allows relatively rapid development from scratch to build applications on top of their integrated ORM + caching solution; on the other hand, it is not a stand-alone caching solution. Gigaspaces is based on Sun's Javaspaces approach, which in turn is based on some academic projects called Linda and Tuplespaces; apparently, they (or one of their customers) built some caching features on top of their Javaspaces model, but I don't know much about it and I've never seen it.
    Our system already have a good O-R mapping layer and looking highly available caching solution for performance and scalability. We may also need some integration with intelligence load-balancing as some of the cache object maybe very large and we want request be routed based on the location of the cache as well as other factors.
    Coherence has all of those features built in, and they have been in use in production applications now for years, so they are well proven. (Many of the systems running Coherence are 24x7x365 systems.)

    We have a wide range of customers, and we would be glad to introduce you to customers with needs that are similar to yours so you could verify that Coherence is the optimal solution for your needs. In the meantime, you can see our online customer stories which now includes Jive Software (as of last Friday). Jive Software provides the forum software that is used by Sun, BEA, IBM, Oracle and JavaLobby for their developer portals, and also for many of the high-volume consumer forums on the Internet.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    Coherence: Shared Memories for J2EE Clusters