[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14079618#comment-14079618
]
Allen Wittenauer commented on HDFS-1094:
With pluggable block placement, I'm not
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12891837#action_12891837
]
Konstantin Shvachko commented on HDFS-1094:
---
We can ignore rack failure, which is
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12889910#action_12889910
]
Scott Carey commented on HDFS-1094:
---
This needs to change p from a constant, to a function
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12889976#action_12889976
]
Rodrigo Schmidt commented on HDFS-1094:
---
Hi Scott,
I totally understand your
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=1289#action_1289
]
Scott Carey commented on HDFS-1094:
---
bq. I think we are more interested in permanent
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12889732#action_12889732
]
Rodrigo Schmidt commented on HDFS-1094:
---
I got permission from Alex Smith to post his
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12887866#action_12887866
]
Tsz Wo (Nicholas), SZE commented on HDFS-1094:
--
Thanks Dhruba for the clear
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12887151#action_12887151
]
dhruba borthakur commented on HDFS-1094:
This is a really an excellent discussion
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886998#action_12886998
]
Rodrigo Schmidt commented on HDFS-1094:
---
I might be wrong, but it seems to me that you
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12887012#action_12887012
]
Joydeep Sen Sarma commented on HDFS-1094:
-
(this is the closest i will get to
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12887061#action_12887061
]
Rodrigo Schmidt commented on HDFS-1094:
---
According to your definition, what prevents
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12887075#action_12887075
]
Joydeep Sen Sarma commented on HDFS-1094:
-
@Rodrigo - it's evident u have not read
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886632#action_12886632
]
Rodrigo Schmidt commented on HDFS-1094:
---
Hairong, would you mind explaining how this
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886768#action_12886768
]
Tsz Wo (Nicholas), SZE commented on HDFS-1094:
--
Here the assumption is,
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886771#action_12886771
]
Joydeep Sen Sarma commented on HDFS-1094:
-
@Rodrigo - how about using a term to
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886783#action_12886783
]
Joydeep Sen Sarma commented on HDFS-1094:
-
regarding T and simulataneous failures:
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886797#action_12886797
]
Scott Carey commented on HDFS-1094:
---
bq. I've had some general discussions with Dhruba
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886823#action_12886823
]
Rodrigo Schmidt commented on HDFS-1094:
---
@Joydeep: Using a term to describe the groups
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886861#action_12886861
]
Joydeep Sen Sarma commented on HDFS-1094:
-
one can't have a different node-group for
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886867#action_12886867
]
Rodrigo Schmidt commented on HDFS-1094:
---
I'm not sure I understand what you mean by
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886900#action_12886900
]
Konstantin Shvachko commented on HDFS-1094:
---
Rodrigo, I understand now that you
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886903#action_12886903
]
Joydeep Sen Sarma commented on HDFS-1094:
-
my mental model is that a node group is a
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886976#action_12886976
]
Joydeep Sen Sarma commented on HDFS-1094:
-
- Pick random rack r2 that is within R
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886979#action_12886979
]
Rodrigo Schmidt commented on HDFS-1094:
---
@Konstantin: We have a script that calculates
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886993#action_12886993
]
Joydeep Sen Sarma commented on HDFS-1094:
-
:-) - of course i meant a node group =
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886254#action_12886254
]
Hairong Kuang commented on HDFS-1094:
-
I did not mean #racks per block. I mean racks for
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886257#action_12886257
]
Rodrigo Schmidt commented on HDFS-1094:
---
As Joydeep wrote, we didn't think this was a
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886363#action_12886363
]
Hairong Kuang commented on HDFS-1094:
-
For a large file, it does matter especially in
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886418#action_12886418
]
Rodrigo Schmidt commented on HDFS-1094:
---
I don't think I understand your use case. It
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886430#action_12886430
]
Konstantin Shvachko commented on HDFS-1094:
---
The math looks good to me (in pdf
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886521#action_12886521
]
Rodrigo Schmidt commented on HDFS-1094:
---
@Konstantin: The new policy will have, for
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886595#action_12886595
]
Hairong Kuang commented on HDFS-1094:
-
I don't think I understand your use case. It
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886599#action_12886599
]
Rodrigo Schmidt commented on HDFS-1094:
---
I think I get your point. I don't think you
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886600#action_12886600
]
Rodrigo Schmidt commented on HDFS-1094:
---
Just complementing my previous comment, I
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886613#action_12886613
]
Hairong Kuang commented on HDFS-1094:
-
Hi Rodrigo, I do not know your algorithm so I
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885844#action_12885844
]
Rodrigo Schmidt commented on HDFS-1094:
---
@Hairong: Just a few observations
1) The
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885845#action_12885845
]
Hairong Kuang commented on HDFS-1094:
-
Right, in case of a map/reduce application,
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885848#action_12885848
]
Hairong Kuang commented on HDFS-1094:
-
Rodrigo, thanks for your explanation. Now I
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885849#action_12885849
]
Rodrigo Schmidt commented on HDFS-1094:
---
The assumption that failures are independent
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885852#action_12885852
]
Joydeep Sen Sarma commented on HDFS-1094:
-
but map-reduce programs rarely read from
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886244#action_12886244
]
Hairong Kuang commented on HDFS-1094:
-
More questions... :)
I understand that the
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12886246#action_12886246
]
Rodrigo Schmidt commented on HDFS-1094:
---
The higher the number of racks, the higher
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885711#action_12885711
]
Hairong Kuang commented on HDFS-1094:
-
This policy seems to decrease the aggregate read
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885765#action_12885765
]
Rodrigo Schmidt commented on HDFS-1094:
---
@Hairong: I'm not sure if I understand your
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885837#action_12885837
]
dhruba borthakur commented on HDFS-1094:
Hi hairong; does your comment about
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12885836#action_12885836
]
Hairong Kuang commented on HDFS-1094:
-
Suppose a file has N blocks. The proposed policy
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12882774#action_12882774
]
Tsz Wo (Nicholas), SZE commented on HDFS-1094:
--
Quoted from prob.pdf:
{quote}
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12882788#action_12882788
]
Rodrigo Schmidt commented on HDFS-1094:
---
There is a number of simplifications there,
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12882794#action_12882794
]
Tsz Wo (Nicholas), SZE commented on HDFS-1094:
--
There is a number of
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12882795#action_12882795
]
Rodrigo Schmidt commented on HDFS-1094:
---
That might make it hard to start reading,
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12882798#action_12882798
]
Koji Noguchi commented on HDFS-1094:
Just curious. Are you getting missing/corrupt
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12878534#action_12878534
]
Rodrigo Schmidt commented on HDFS-1094:
---
For this JIRA I'm implementing a general
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12873238#action_12873238
]
dhruba borthakur commented on HDFS-1094:
From all the above mathematical formulae,
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856510#action_12856510
]
Karthik Ranganathan commented on HDFS-1094:
---
Hey Brian,
Yes, that would help with
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856673#action_12856673
]
Konstantin Shvachko commented on HDFS-1094:
---
P(of not losing a specific block) =
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856716#action_12856716
]
Jitendra Nath Pandey commented on HDFS-1094:
I think you are throwing one
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12855883#action_12855883
]
dhruba borthakur commented on HDFS-1094:
This is more of a discussion on how to
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856139#action_12856139
]
Brian Bockelman commented on HDFS-1094:
---
Hey Karthik,
Let me play dumb (it might not
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856223#action_12856223
]
Brian Bockelman commented on HDFS-1094:
---
A funny related anecdote that I've heard
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856269#action_12856269
]
Rodrigo Schmidt commented on HDFS-1094:
---
Yes, if we want to minimize the chances of
[
https://issues.apache.org/jira/browse/HDFS-1094?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=12856271#action_12856271
]
Brian Bockelman commented on HDFS-1094:
---
Hey Karthik,
Why don't we take this to the
61 matches
Mail list logo