AFAIR the max vector chunk sizes come from the xrootd protocol, and the HTTP code tries to fit the more liberal HTTP specs into that, limiting the overhead. I am puzzled by the idea to work around the vector size limitations imposed (and well documented) by the xrootd protocol
Which client is generating these requests? I don't think that a vector read makes too much sense for large chunks
That's correct, that's why I'm asking what is wrong with the algorithm : its behavior, or its shape? Thanks for your comment @ffurano !
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
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