1 |
On Mon, 17 Dec 2018 19:57:21 -0800 |
2 |
Georgy Yakovlev <gyakovlev@g.o> wrote: |
3 |
|
4 |
> I'm guessing, but probably infra server is not supposed to handle load from |
5 |
> all the users and will temporarily ban if one tries to sync more than several |
6 |
> times per day (like rsync master does). But don't quote me on that, better ask |
7 |
> infra. |
8 |
|
9 |
I'd imagine the server requirements with regard to load, is less for |
10 |
git than it is for rsync. |
11 |
|
12 |
Partly, because I believe rsync's require tree traversal, and dynamic |
13 |
checksumming of data on the server side for each sync. |
14 |
|
15 |
Whereas with Git, that checksumming and traversal are essentially |
16 |
precomputed, and the backing store can be efficiently condensed to a |
17 |
single file, with much more efficient IO. |
18 |
|
19 |
That is, instead of iterating through 9k+ inodes, it just opens the |
20 |
one and chases the parent SHA1 chains. |
21 |
|
22 |
Then your restrictions seem to amount to total bandwidth available, |
23 |
with a little CPU and IO overhead, as opposed to a larger bandwith, CPU |
24 |
and IO requirement. |