1 / 9

RPS Notes and RPS/REMOS Relational Schema

RPS Notes and RPS/REMOS Relational Schema. Peter A. Dinda. RPS Toolkit. Extensible toolkit for implementing resource signal prediction systems Easy “buy-in” for users C++ and sockets (no threads) Prebuilt prediction components Libraries (sensors, time series, communication)

grainger
Télécharger la présentation

RPS Notes and RPS/REMOS Relational Schema

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. RPS Notes and RPS/REMOS Relational Schema Peter A. Dinda

  2. RPS Toolkit • Extensible toolkit for implementing resource signal prediction systems • Easy “buy-in” for users • C++ and sockets (no threads) • Prebuilt prediction components • Libraries (sensors, time series, communication) • Users have bought in • Incorporated in CMU Remos, BBN QuO • Various research users http://www.cs.northwestern.edu/~pdinda/RPS.html

  3. Conceptual Model of an RPS-based System RPS components can be composed in other ways

  4. Common Configuration

  5. Queries and Data Models for Prediction and MeasurementIn RPS and Remos Peter A. Dinda Remos Meeting 3/9/99

  6. Measurement/Prediction Queries • “Where can I find a load measurement stream for the host firenze?” • source:tcp:firenze:5000, source:udp:239.99.99.99:5000 • “Give me the latest prediction server implementation for ALPHA DUX 3.2” • blob(executable) • “Where can I find a prediction buffer for the underlying load measurement stream “firenze””? • requestresponse:tcp:pyramid:8193

  7. Resource Queries • “I need 1 to 4 machines, all the same OS and arch, with a combined memory of 1 GB” • (mojave),(sahara),((poconos,pyramid,foo),(manch1,2,3,4), etc) • ...

  8. Database Schema benchmarks BT numproc mhz arch os osv mem vmem dasd perf perfblob note UR hostdata ip numproc mhz arch os osv mem vmem dasd loc user note UR hosts modules ip name mid mt dsid note datasources moduleexecs dsid dst mt arch os minosv ver execblob note UR flowssources endpoints dsid ip1 ip2 mid epid nodesource endpointdata dsid ip epid ct ip port fn note UR

  9. Query Example [pdinda@firenze ~]$ find_hosts_mem.pl find_hosts_mem.pl minmem maxmem minnum maxnum limit [samearchos] [pdinda@firenze ~]$ find_hosts_mem.pl 512 1024 1 3 5 1 Host solutions select host1.name, hd1.arch, hd1.os, hd1.mem as TotalMem from hosts as host1, hostdata as hd1 where host1.ip=hd1.ip and hd1.mem >= 512 and hd1.mem <= 1024 order by TotalMem desc limit 5 +------------------------+-------+-----+----------+ | name | arch | os | TotalMem | +------------------------+-------+-----+----------+ | mojave.cmcl.cs.cmu.edu | ALPHA | DUX | 1024.00 | | sahara.cmcl.cs.cmu.edu | ALPHA | DUX | 1024.00 | +------------------------+-------+-----+----------+ 2 Host solutions select host1.name, hd1.arch, hd1.os, host2.name, hd2.arch, hd2.os, hd1.mem+ hd2.mem as TotalMem from hosts as host1, hostdata as hd1, hosts as host2, hostdata as hd2 where host1.ip=hd1.ip and host2.ip=hd2.ip and hd1.mem + hd2.mem >= 512 and hd1.mem + hd2.mem <= 1024 and host1.ip != host2.ip order by TotalMem desc limit 5 +---------------------------+-------+-------+---------------------------+-------+-------+----------+ | name | arch | os | name | arch | os | TotalMem | +---------------------------+-------+-------+---------------------------+-------+-------+----------+ | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | 512.00 | | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | INNUENDO.TRUST.CS.CMU.EDU | I386 | LINUX | 512.00 | | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | pyramid.cmcl.cs.cmu.edu | ALPHA | DUX | 512.00 | | firenze.cmcl.cs.cmu.edu | I386 | LINUX | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | 512.00 | | INNUENDO.TRUST.CS.CMU.EDU | I386 | LINUX | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | 512.00 | +---------------------------+-------+-------+---------------------------+-------+-------+----------+ 3 Host solutions select host1.name, hd1.arch, hd1.os, host2.name, hd2.arch, hd2.os, host3.name, hd3.arch, hd3.os, hd1.mem+ hd2.mem+ hd3.mem as TotalMem from hosts as host1, hostdata as hd1, hosts as host2, hostdata as hd2, hosts as host3, hostdata as hd3 where host1.ip=hd1.ip and host2.ip=hd2.ip and host3.ip=hd3.ip and hd1.mem + hd2.mem + hd3.mem >= 512 and hd1.mem + hd2.mem + hd3.mem <= 1024 and host1.ip != host2.ip and host1.ip != host3.ip and host2.ip != host3.ip order by TotalMem desc limit 5 +---------------------------+-------+-------+---------------------------+-------+-------+---------------------------+------+-------+----------+ | name | arch | os | name | arch | os | name | arch | os | TotalMem | +---------------------------+-------+-------+---------------------------+-------+-------+---------------------------+------+-------+----------+ | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | INNUENDO.TRUST.CS.CMU.EDU | I386 | LINUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | 640.00 | | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | pyramid.cmcl.cs.cmu.edu | ALPHA | DUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | 640.00 | | INNUENDO.TRUST.CS.CMU.EDU | I386 | LINUX | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | 640.00 | | pyramid.cmcl.cs.cmu.edu | ALPHA | DUX | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | 640.00 | | poconos.cmcl.cs.cmu.edu | ALPHA | DUX | firenze.cmcl.cs.cmu.edu | I386 | LINUX | INNUENDO.TRUST.CS.CMU.EDU | I386 | LINUX | 640.00 | +---------------------------+-------+-------+---------------------------+-------+-------+---------------------------+------+-------+----------+

More Related