Hi Raja,
Indeed, Ceph reads the block multiple times, one for each call to read
that block. As for experience needed, alsmost any good C++ developer could
do this. The code is enacsulated in a plugin with little xrootd
entanglement. Likely, they would need to know more about Ceph than xrootd.
Andy
On Tue, 18 Aug 2020, Raja Nandakumar wrote:
> Hi @osschar ,
>
> Thanks a lot for the two detailed analyses. A few questions :
>
> 1. In the case of whole reads, does CEPH still have to read a block >>1 times to get the needed data out?
> 2. In any case, If I understand correct, fixing the vector reads on the xrootd-ceph can only help. I am guessing that this will be done by someone with xrootd knowledge, and I would be happy to do my best to test it at the RAL end if possible / needed. A timescale for this would be useful.
> 3. We are also investigating one set of WNs here at RAL which seem to have multiple issues (ncluding these failures). I will let you know if we find anything about it.
>
> Thanks and Cheers,
> Raja.
>
> --
> You are receiving this because you were mentioned.
> Reply to this email directly or view it on GitHub:
> https://github.com/xrootd/xrootd/issues/1259#issuecomment-675737000
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or unsubscribe.
[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"target": "https://github.com/xrootd/xrootd/issues/1259#issuecomment-675743509",
"url": "https://github.com/xrootd/xrootd/issues/1259#issuecomment-675743509",
"name": "View Issue"
},
"description": "View this Issue on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1