Re: [galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
Glad to see someone else is playing around with Mesos. I have a mesos branch that is getting a little long in the tooth. I'd like to get a straight job runner (non-LWR, with a shared file system) running under mesos for Galaxy before I submit that work for a pull request. The hackathon is only 12 days away! Hopefully we'll be able to make some progress on these sorts of projects. Kyle On Sun, Jun 15, 2014 at 4:06 PM, John Chilton jmchil...@gmail.com wrote: Hey Kyle, all, If anyone wants to play with running Galaxy jobs within an Apache Mesos environment I have added a prototype of this feature to the LWR. https://bitbucket.org/jmchilton/lwr/commits/555438d2fe266899338474b25c540fef42bcece7 https://bitbucket.org/jmchilton/lwr/commits/9748b3035dbe3802d4136a6a1028df8395a9aeb3 This work distributes jobs across a Mesos cluster and injects a MESOS_URL environment variable into the job runtime environment in case the jobs themselves want to take advantage of Mesos. The advantage of the LWR versus a traditional Galaxy runner is that the job can be staged to remote resources without shared disk. Prior to this I was imaging the LWR to be useful in cases where Galaxy and remote cluster don't share common disk but where there is in fact a shared scratch directory or something across the remote cluster as well a resource manager. The LWR Mesos framework however has the actual compute servers themselves stage the job up and down - so you could imagine distributing Galaxy across large clusters without any shared disk whatsoever - that could be very cool and help scale say cloud applications. Downsides of an LWR-based approach versus a Galaxy approach is that it is less mature and there is more stuff to configure - need to configure a Galaxy job_conf plugin and destination, need to configure the LWR itself, need to configure a message queue (for this variant of LWR operation anyway - it should be possible to drive this via the LWR in web server mode but I haven't added it yet). I would be more than happy to continue to see progress toward Mesos support in Galaxy proper. It is strictly a prototype so far - a sort of playground if anyone wants to play with these ideas and build something cool. It really is a framework right - not so much a job scheduler so I am not sure it is very immediately useful - but I imagine one could build cool stuff on top of it. Next, I think I would like to add Apache Aurora (http://aurora.incubator.apache.org/) support - because it seems like a much more traditional resource manager but built on top of Mesos so it would be more practical for traditional Galaxy-style jobs. Doesn't buy you anything in terms of parallelization but it would fit better with Galaxy. -John On Sat, Oct 26, 2013 at 2:43 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/ ). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI ( https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md ) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is then passed to tool wrappers and allows them to contact the local mesos infrastructure (assuming the system has been configured) or pass a null if the system isn't available. 2) Write a tool runner that works as a mesos framework to executes single cpu jobs on the distributed system. 3) For instances where mesos is not available at a system wide level (say they only have access to an SGE based cluster), but the user wants to run distributed jobs, write a wrapper that can create a mesos cluster using the existing queueing system. For example, right now I run a Mesos system under the SGE queue system. I'm curious to see what other people think. Kyle ___ Please keep all replies on the list by using reply all in your
Re: [galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
Hey Kyle, all, If anyone wants to play with running Galaxy jobs within an Apache Mesos environment I have added a prototype of this feature to the LWR. https://bitbucket.org/jmchilton/lwr/commits/555438d2fe266899338474b25c540fef42bcece7 https://bitbucket.org/jmchilton/lwr/commits/9748b3035dbe3802d4136a6a1028df8395a9aeb3 This work distributes jobs across a Mesos cluster and injects a MESOS_URL environment variable into the job runtime environment in case the jobs themselves want to take advantage of Mesos. The advantage of the LWR versus a traditional Galaxy runner is that the job can be staged to remote resources without shared disk. Prior to this I was imaging the LWR to be useful in cases where Galaxy and remote cluster don't share common disk but where there is in fact a shared scratch directory or something across the remote cluster as well a resource manager. The LWR Mesos framework however has the actual compute servers themselves stage the job up and down - so you could imagine distributing Galaxy across large clusters without any shared disk whatsoever - that could be very cool and help scale say cloud applications. Downsides of an LWR-based approach versus a Galaxy approach is that it is less mature and there is more stuff to configure - need to configure a Galaxy job_conf plugin and destination, need to configure the LWR itself, need to configure a message queue (for this variant of LWR operation anyway - it should be possible to drive this via the LWR in web server mode but I haven't added it yet). I would be more than happy to continue to see progress toward Mesos support in Galaxy proper. It is strictly a prototype so far - a sort of playground if anyone wants to play with these ideas and build something cool. It really is a framework right - not so much a job scheduler so I am not sure it is very immediately useful - but I imagine one could build cool stuff on top of it. Next, I think I would like to add Apache Aurora (http://aurora.incubator.apache.org/) support - because it seems like a much more traditional resource manager but built on top of Mesos so it would be more practical for traditional Galaxy-style jobs. Doesn't buy you anything in terms of parallelization but it would fit better with Galaxy. -John On Sat, Oct 26, 2013 at 2:43 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/ ). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI (https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is then passed to tool wrappers and allows them to contact the local mesos infrastructure (assuming the system has been configured) or pass a null if the system isn't available. 2) Write a tool runner that works as a mesos framework to executes single cpu jobs on the distributed system. 3) For instances where mesos is not available at a system wide level (say they only have access to an SGE based cluster), but the user wants to run distributed jobs, write a wrapper that can create a mesos cluster using the existing queueing system. For example, right now I run a Mesos system under the SGE queue system. I'm curious to see what other people think. Kyle ___ Please keep all replies on the list by using reply all in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/ ___ Please keep all replies on the list by using reply all in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at:
Re: [galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
Hi Kyle, Swift indeed is a complete framework for distributed computing. Distributing files out to cluster nodes, starting processes, bringing back result files to submit host is done out of the box (stagein-exec-stageout cycle). We can discuss offline if you are interested in giving it a shot. Best, Ketan On Mon, Oct 28, 2013 at 4:14 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: You probably are a good person to get an opinion from. My plan isn't to write new frameworks, but rather use existing libraries that can communicate with Mesos to setup their parallel environments. But for Swift, you would probably want to write a new framework. Just looking at Swift, I imagine one of the harder parts is just getting the system setup on a cluster (ie distributing out files to remote nodes, making sure that you have a way to start processes on those nodes and have them know where to find the master), it seems like Swift could benefit from having a Mesos based framework. Do you think it would enable you to have a 'zero-config' startup of a distributed Swift application? Kyle On Mon, Oct 28, 2013 at 1:51 PM, Ketan Maheshwari ketancmaheshw...@gmail.com wrote: Hi Kyle, We have a similar ongoing development wherein we are working on integrating our Swift framework ( swift-lang.org ) with Galaxy. The goal is to enable Galaxy based applications to run on a variety of distributed resources via various integration schemes as suitable to application and underlying execution environment. Here is an abstract of a paper (co-authored with Ravi, who responded on this thread) we will be presenting in a workshop at the upcoming SC 13 conference: The Galaxy platform is a web-based science portal for scientific computing supporting Life Sciences users community. While user-friendly and intuitive for doing small to medium scale computations, it currently has a limited support for large-scale, parallel and distributed computing. The Swift parallel scripting framework is capable of composing ordinary applications into parallel scripts that can be run on multi-scale distributed and performance computing platforms. In complex distributed environments, often the user end of application lifecycle slows down because of the technical complexities brought in by the scale, access methods and resource management nuances. Galaxy offers a simple way of designing, composing, executing, reusing, and reproducing application runs. An integration between Swift and Galaxy systems can accelerate science as well as bring the respective user communities together in an interactive, user-friendly, parallel and distributed data analysis environment enabled on a broad range of computational infrastructures. Kindly let us know if you need a hands on for the various tools we have already developed. Best, Ketan On Mon, Oct 28, 2013 at 3:07 PM, Kyle Ellrott kellr...@soe.ucsc.eduwrote: I don't think implementation will be very difficult. The bigger question is this a technology people are open to? The nearest competitor is YARN ( http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/YARN.html). Mesos seems a bit more geared toward general purpose usage (with several existing frameworks), while YARN seems more specific to Hadoop. But I'd be glad to hear some other thoughts. Kyle On Mon, Oct 28, 2013 at 12:55 PM, Ravi K Madduri madd...@mcs.anl.govwrote: Kyle This is something I am very interested in. The three parts below make sense to me. I would be very happy to discuss further and provide any help to move this forward. Regards On Oct 26, 2013, at 2:43 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/ ). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI ( https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md ) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is
Re: [galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
I don't think implementation will be very difficult. The bigger question is this a technology people are open to? The nearest competitor is YARN ( http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/YARN.html). Mesos seems a bit more geared toward general purpose usage (with several existing frameworks), while YARN seems more specific to Hadoop. But I'd be glad to hear some other thoughts. Kyle On Mon, Oct 28, 2013 at 12:55 PM, Ravi K Madduri madd...@mcs.anl.govwrote: Kyle This is something I am very interested in. The three parts below make sense to me. I would be very happy to discuss further and provide any help to move this forward. Regards On Oct 26, 2013, at 2:43 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI ( https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md ) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is then passed to tool wrappers and allows them to contact the local mesos infrastructure (assuming the system has been configured) or pass a null if the system isn't available. 2) Write a tool runner that works as a mesos framework to executes single cpu jobs on the distributed system. 3) For instances where mesos is not available at a system wide level (say they only have access to an SGE based cluster), but the user wants to run distributed jobs, write a wrapper that can create a mesos cluster using the existing queueing system. For example, right now I run a Mesos system under the SGE queue system. I'm curious to see what other people think. Kyle ___ Please keep all replies on the list by using reply all in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/ -- Ravi K Madduri MCS, Argonne National Laboratory Computation Institute, University of Chicago ___ Please keep all replies on the list by using reply all in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
Re: [galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
You probably are a good person to get an opinion from. My plan isn't to write new frameworks, but rather use existing libraries that can communicate with Mesos to setup their parallel environments. But for Swift, you would probably want to write a new framework. Just looking at Swift, I imagine one of the harder parts is just getting the system setup on a cluster (ie distributing out files to remote nodes, making sure that you have a way to start processes on those nodes and have them know where to find the master), it seems like Swift could benefit from having a Mesos based framework. Do you think it would enable you to have a 'zero-config' startup of a distributed Swift application? Kyle On Mon, Oct 28, 2013 at 1:51 PM, Ketan Maheshwari ketancmaheshw...@gmail.com wrote: Hi Kyle, We have a similar ongoing development wherein we are working on integrating our Swift framework ( swift-lang.org ) with Galaxy. The goal is to enable Galaxy based applications to run on a variety of distributed resources via various integration schemes as suitable to application and underlying execution environment. Here is an abstract of a paper (co-authored with Ravi, who responded on this thread) we will be presenting in a workshop at the upcoming SC 13 conference: The Galaxy platform is a web-based science portal for scientific computing supporting Life Sciences users community. While user-friendly and intuitive for doing small to medium scale computations, it currently has a limited support for large-scale, parallel and distributed computing. The Swift parallel scripting framework is capable of composing ordinary applications into parallel scripts that can be run on multi-scale distributed and performance computing platforms. In complex distributed environments, often the user end of application lifecycle slows down because of the technical complexities brought in by the scale, access methods and resource management nuances. Galaxy offers a simple way of designing, composing, executing, reusing, and reproducing application runs. An integration between Swift and Galaxy systems can accelerate science as well as bring the respective user communities together in an interactive, user-friendly, parallel and distributed data analysis environment enabled on a broad range of computational infrastructures. Kindly let us know if you need a hands on for the various tools we have already developed. Best, Ketan On Mon, Oct 28, 2013 at 3:07 PM, Kyle Ellrott kellr...@soe.ucsc.eduwrote: I don't think implementation will be very difficult. The bigger question is this a technology people are open to? The nearest competitor is YARN ( http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/YARN.html). Mesos seems a bit more geared toward general purpose usage (with several existing frameworks), while YARN seems more specific to Hadoop. But I'd be glad to hear some other thoughts. Kyle On Mon, Oct 28, 2013 at 12:55 PM, Ravi K Madduri madd...@mcs.anl.govwrote: Kyle This is something I am very interested in. The three parts below make sense to me. I would be very happy to discuss further and provide any help to move this forward. Regards On Oct 26, 2013, at 2:43 PM, Kyle Ellrott kellr...@soe.ucsc.edu wrote: I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/ ). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI ( https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md ) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is then passed to tool wrappers and allows them to contact the local mesos infrastructure (assuming the system has been configured) or pass a null if the system isn't available. 2) Write a tool runner that works as a mesos framework to executes single cpu jobs on the distributed system. 3) For instances where mesos is not available at a system wide level (say they only have access to an SGE based
[galaxy-dev] Using Mesos to Enable distributed computing under Galaxy?
I think one of the aspects where Galaxy is a bit soft is the ability to do distributed tasks. The current system of split/replicate/merge tasks based on file type is a bit limited and hard for tool developers to expand upon. Distributed computing is a non-trival thing to implement and I think it would be a better use of our time to use an already existing framework. And it would also mean one less API for tool writers to have to develop for. I was wondering if anybody has looked at Mesos ( http://mesos.apache.org/). You can see an overview of the Mesos architecture at https://github.com/apache/mesos/blob/master/docs/Mesos-Architecture.md The important thing about Mesos is that it provides an API for C/C++, Java/Scala and Python to write distributed frameworks. There are already implementations of frameworks for common parallel programming systems such as: - Hadoop (https://github.com/mesos/hadoop) - MPI ( https://github.com/apache/mesos/blob/master/docs/Running-torque-or-mpi-on-mesos.md ) - Spark (http://spark-project.org) And you can find example Python framework at https://github.com/apache/mesos/tree/master/src/examples/python Integration with Galaxy would have three parts: 1) Add a system config variable to Galaxy called 'MESOS_URL' that is then passed to tool wrappers and allows them to contact the local mesos infrastructure (assuming the system has been configured) or pass a null if the system isn't available. 2) Write a tool runner that works as a mesos framework to executes single cpu jobs on the distributed system. 3) For instances where mesos is not available at a system wide level (say they only have access to an SGE based cluster), but the user wants to run distributed jobs, write a wrapper that can create a mesos cluster using the existing queueing system. For example, right now I run a Mesos system under the SGE queue system. I'm curious to see what other people think. Kyle ___ Please keep all replies on the list by using reply all in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/