BT

Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ

Topics

Choose your language

InfoQ Homepage Presentations The Future of the JVM

The Future of the JVM

Bookmarks
01:10:14

Summary

The panelists discuss the future of the JVM in the context of parallelism and high concurrency of tomorrow’s thousands of cores.

Bio

Jamie Allen - authoring, Effective Akka Cliff Click - CTO & Founder, 0xdata Charlie Hunt - Author, Java Performance Doug Lea - Governing Board, OpenJDK Michael Pilquist - Lead Software Architect, CCAD

About the conference

The ETE Conference has established itself as the most diverse and interesting conference on the East Coast. Curated by developers, for developers, it brings together the brightest minds in software development. Visit emergingtech.chariotsolutions.com we provide content throughout the year and you can subscribe to our Chariot TechCast podcast.

Hello stranger!

You need to Register an InfoQ account or or login to post comments. But there's so much more behind being registered.

Get the most out of the InfoQ experience.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Community comments

  • Politics May Kill the Java / JVM if Innovation does not rule.

    by Suminda Dharmasena,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    Hi,

    From what I gather the politics is slowing down the JVM innovation and politics consumes more resources than what is put to innovate as some trivial things are not put into the JVM where politics take an upper hand. Starting point is to get politics out or to a minimal. If this is not tackled then it might take 10 year for innovation which can be rolled out in the next version. What we might see in Java 8 and then Java 9 is what should have been Java 7.

    I feels this should be dealt with in the next version:
    - 1 year coordinated release cycle at a particular time of year like Eclipse
    - value types
    - reified generics along side erasure based generics
    - multi dimensional arrays
    - box elimination (value type primitive objects)
    - less GC fallback (better stack allocation, release objects on last reference if it does not escape - if containing objects live longer delete the ones that can be and leave the rest for GC)
    - pluggable GC / memory management
    - Platform independent core class for Unsafe like operations without using Unsafe. Perhaps Grizzly like release.
    - low level High Performance Computing like constructs
    - separate JVM from Java
    - modulations so we can have components mixed and matched from different profiles

    Suminda

  • Politics May Kill the Java / JVM if Innovation does not rule.

    by Suminda Dharmasena,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    Hi,

    From what I gather the politics is slowing down the JVM innovation and politics consumes more resources than what is put to innovate as some trivial things are not put into the JVM where politics take an upper hand. Starting point is to get politics out or to a minimal. If this is not tackled then it might take 10 year for innovation which can be rolled out in the next version. What we might see in Java 8 and then Java 9 is what should have been Java 7.

    I feels this should be dealt with in the next version:
    - 1 year coordinated release cycle at a particular time of year like Eclipse
    - value types
    - reified generics along side erasure based generics
    - multi dimensional arrays
    - box elimination (value type primitive objects)
    - less GC fallback (better stack allocation, release objects on last reference if it does not escape - if containing objects live longer delete the ones that can be and leave the rest for GC)
    - pluggable GC / memory management
    - Platform independent core class for Unsafe like operations without using Unsafe perhaps using a Memory Manager. Perhaps Grizzly like allocate, reallocate and release for general usage.
    - easier use of thread local allocation with compiler / JVM mediation (also @ThreadLocal)
    - low level High Performance Computing like constructs
    - separate JVM from Java
    - modulations so we can have components mixed and matched from different profiles
    - C4 collector
    - Weakreference through tags e.g. @Weak, @Soft
    - Also leverage from other company innovation like IBM J9

    Suminda

  • JVM Cleanup

    by Suminda Dharmasena,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    A JVM cleanup is also much warranted. If it is the legacy perhaps clean the libs and JVM. The provide a compatibility layer and also a migration tool to gen wrappers around few functionality. Complexity of legacy can be a killer.

    Perhaps a middle road than a complete clean slate to more massive overhaul retaining only well designed and useful features.

    Also better efficiency and effectiveness.

  • JVM Cleanup

    by Suminda Dharmasena,

    Your message is awaiting moderation. Thank you for participating in the discussion.

    A JVM cleanup is also much warranted. If it is the legacy perhaps clean the libs and JVM. The provide a compatibility layer and also a migration tool to gen wrappers around few functionality. Complexity of legacy can be a killer.

    Perhaps a middle road than a complete clean slate to more massive overhaul retaining only well designed and useful features.

    Also better efficiency and effectiveness.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

BT