Log of the #duraspace-ff channel on chat.freenode.net

Using timezone: Eastern Standard Time
* ksclarke joins01:22
* ksclarke leaves01:25
* fcrepo-bot joins06:35
* fcrepo-bot leaves07:38
* benpennell2 joins07:58
* daines joins08:56
* tecoripa joins09:03
* fasseg joins09:37
<pivotal-bot____>Frank Asseg added comment: "with 10mb datatstreams and ten threads on a six node cluster I had about 6.49ms/s throughput.09:48
Im running a t..." https://www.pivotaltracker.com/story/show/60557096
Frank Asseg started "Clustered F4 Ingest Benchmarks" https://www.pivotaltracker.com/story/show/60557096
Frank Asseg started "Test different confguration settings of infinispan/jgroups in order to raise performance on the FIZ cluster" https://www.pivotaltracker.com/story/show/5940135009:49
Gregory Jansen started "Establish performance goals for UNC cluster" https://www.pivotaltracker.com/story/show/6550482410:18
Gregory Jansen started "Demonstrate that we can predictably scale up a cluster to meet performance goals." https://www.pivotaltracker.com/story/show/65505154
Gregory Jansen deleted "Switch to embedded tomcat7 in fcrepo-test-profiles for cluster test harness." https://www.pivotaltracker.com/story/show/65505672
* scossu joins10:40
* dwilcox joins10:56
<pivotal-bot____>Mike Daines added comment: "- This is partially met in that objects nodes could be associated with license nodes, and license nodes asso..." https://www.pivotaltracker.com/story/show/6629923611:19
Mike Daines added comment: "- Partially met by RBAC reference implementation, but what about groups from LDAP? Authorization documentati..." https://www.pivotaltracker.com/story/show/6629918611:20
Mike Daines added comment: "- Met by offering an API for updating ACLs, and writing datastreams" https://www.pivotaltracker.com/story/show/66299186
Mike Daines added comment: "- Partially met by RBAC reference implementation, but what about groups from LDAP? Authorization documentati..." https://www.pivotaltracker.com/story/show/66299136
Mike Daines added comment: "- Presumably, this could be met through inheritance in RBAC implementation." https://www.pivotaltracker.com/story/show/6629891211:21
Mike Daines added comment: "- Could be met by implementing FedoraPolicyEnforcementPoint?
- Currently have XACML policies at UNC, but in ..." https://www.pivotaltracker.com/story/show/66298796
Mike Daines added comment: "- In UNC's case, this boils down to allowing certain clients to forward HTTP headers11:22
- Met by consumer imple..." https://www.pivotaltracker.com/story/show/66298776
Mike Daines added comment: "- Met by RBAC implementation offering inheritance and the ability of those who have the writer role on an AC..." https://www.pivotaltracker.com/story/show/66298724
Mike Daines added comment: "- Possibly would be met by client implementation of PEP interface." https://www.pivotaltracker.com/story/show/6629875411:23
* ksclarke joins11:27
Is there a max file size for fedora 3?
<benpennell2>never heard of there being one, but if there is its very large11:38
<ksclarke>thanks!11:39
<benpennell2>all, here is the performance versus single node repo size data i mentioned in the call:11:40
https://docs.google.com/spreadsheet/ccc?key=0Akmxs4dcuisQdG1jUW1JZ3ltUHYzclhoWUhpZHV4VGc&usp=sharing
<daines>What are the units for "throughput" in the lower graph?11:45
<benpennell2>MB/sec according to benchtool11:46
i'll add that in
<daines>In the first graph, is the horizontal axis the number of datastreams which are 200kb large? or does 40000 mean 40000*200kb?11:48
<benpennell2>number of datastreams, which are 200kb large each
<daines>It seems like the ingest time varies more as there are more datastreams?11:49
<benpennell2>yes, the swings were getting larger from run to run the closer the server came to the end11:50
<daines>But once the number of datastreams gets to 20000, the throughput is more or less constant?
(Maybe)11:51
<benpennell2>that is curious, it is still declining, but i guess with the way its calculated the number is already so small that further decrease doesn't show up well11:52
<daines>I guess the question is: how slowly is it declining?11:53
I have posted suggestions for possible modifications to the AuthZ interfaces as a child of the authn/authz design page: https://wiki.duraspace.org/display/FF/Possible+modifications+to+AuthZ+Interfaces11:56
<pivotal-bot____>Mike Daines accepted "Review AuthZ Use Cases (Mike)" https://www.pivotaltracker.com/story/show/66298620
Mike Daines started "Remove filterPathsForReading from FedoraPolicyEnforcementPoint interface" https://www.pivotaltracker.com/story/show/65657140
* awoods joins
<benpennell2>he returns11:58
<awoods>benpennell2: me?11:59
<benpennell2>you
<awoods>benpennell2: It is good to be back... although this week will be a bit fragmented.
<fasseg>re awoods :) hope you had a nice time teambuilding12:07
<awoods>fasseg: it was very nice indeed.
<fasseg>and walking straight isnt a problem anymore? ;)12:08
* aawoods_ joins12:31
* aawoods_ leaves
* awoods leaves12:32
* aawoods_ joins
* benpennell2 leaves12:40
* ksternad joins12:57
* github-ff joins12:58
[fcrepo4] daines opened pull request #261: Remove filterPathsForReading from FedoraPolicyEnforcement (master...remove-filter-paths) http://git.io/n6HxuA
* github-ff leaves
<pivotal-bot____>Mike Daines added comment: "Addressed in https://github.com/futures/fcrepo4/pull/261" https://www.pivotaltracker.com/story/show/65657140
Mike Daines finished "Remove filterPathsForReading from FedoraPolicyEnforcementPoint interface" https://www.pivotaltracker.com/story/show/65657140
* benpennell2 joins
* edInCo joins13:10
* tecoripa leaves13:35
<awoods>Sprint Team: Any specific news from the sprint last week?14:38
* ksternad leaves14:51
* fcrepo-bot joins14:57
* scossu leaves15:40
* dwilcox leaves15:57
* fcrepo-bot leaves15:59
* scossu joins16:04
* scossu1 joins17:40
* scossu leaves17:42
* benpennell2 leaves18:03
* edInCo leaves18:16
* awoods leaves18:18
<pivotal-bot____>Andrew Woods accepted "Cache Federated SHA-1 values" https://www.pivotaltracker.com/story/show/6522232018:31
* awoods joins18:32
* ksclarke leaves18:40
* benpennell2 joins19:01
* tecoripa joins19:04
* tecoripa leaves
* scossu1 leaves19:53
* fasseg leaves20:42
* scossu joins21:10
* scossu leaves21:15
* ksclarke joins22:14
* ksclarke leaves23:15
* benpennell2 leaves23:26

Generated by Sualtam