[cloud-workshop] Notes from OSGi Cloud Workshop at Eclipsecon
david at redhat.com
Thu Mar 29 17:05:38 EDT 2012
Thanks for joining the discussion today. I thought it was a highly
productive and enjoyable morning.
I took some notes of the topics that were brought forward during the
meeting. Please find them below. Do let me know if I captured something
incorrectly or something is missing.
Also, if you're interested in taking part in the further technical
design of any of these, do let me and/or co-chair Tim Diekmann (tdiekman
at tibco.com) know and we will fill you in on how to get involved in the
OSGi EEG co-chair
BTW I'm planning to collect the presentations that were given here:
Below you can find the topics that were proposed today as areas to work
on in the OSGi Enterprise Expert Group in the context of Cloud
Computing. The list is not in any particular order.
Topic: Make it possible to describe various Service unavailable States.
A service may be unavailable in the cloud because of a variety of reasons.
Maybe the amount of invocations available to you have exhausted for today.
Maybe your credit card expired
Maybe the node running the service crashed.
It should be possible to model these various failure states and it
should also be possible to register 'callback' mechanisms that can deal
with these states in whatever way is appropriate (blacklist the service,
wait a while, send an email to the credit card holder, etc).
Topic: WYTIWYR (What you test is what you run) It should be possible to
quickly deploy and redeploy.
There was an additional use-case around reverting the data (and
configuration) changes made during an upgrade. If we need to downgrade
after an upgrade then we may need to convert the data/configuration back
into its old state. Peter Kriens says that there had been a lot of
discussion about reverting deployments in the past and that it was
decided not to support it.
Stefan Derdak: you need to be careful of feature creep in the framework.
This seems to outside of the realm of the framework.
David Bosschaert: possibly it can already be addressed with the current
building blocks. Maybe a whitepaper could help?
Topic: Come up with a common and agreed architecture for Discovery. This
should include consideration of Remote Services, Remote Events and
Distributed Configuration Admin.
Topic: Using OSGi in the cloud should be super easy and frictionless.
Topic: Resource utilization. It should be possible to measure/report
this for each cloud node. Number of threads available, amount of memory,
power consumption etc… Possibly create OSGi Capability namespaces for this.
Topic: OBR scaling. Need to be able to use OBR in a highly available
manner. Should support failover and should hook in with discovery.
Topic: We need subsystems across frameworks. Possibly refer to them as
'Ecosystems'. These describe a number of subsystems deployed across a
number of frameworks.
Topic: Asynchronous services and asynchronous remote services.
Topic: Isolation and security for applications
Protect access to file system
Lifecycle handling of applications
OBR - isolated OBR (multiple tenants should not see each other's OBR)
This all needs to be configurable
Topic: It should be possible to look at the cloud system state:
where am I (type of cloud, geographical location)?
what nodes are there and what is their state?
what frameworks are available in this cloud system?
where's my OBR?
what state am I in?
what do I need here in order to operate?
Topic: There should be a management mechanism for use in the cloud
JMX? Possibly not
REST? Most likely
Management of application state should also be possible in addition to
Topic: Deployment - when deploying replicated nodes it should be
possible to specify that the replica should *not* be deployed on certain
nodes, to avoid that all the replicas are deployed on the same node.
Topic: Review all OSGi service specifications for suitability in the cloud.
More information about the cloud-workshop