b46636df3a
Change-Id: Ibddfb5a97e1ec30dd652d5451698af68936e237b
18 lines
959 B
Plaintext
18 lines
959 B
Plaintext
I intend to continue serving as Swift PTL.
|
|
|
|
I was going to attempt to enumerate some sort of grand vision for Swift and our
|
|
priorities for the next cycle. As I thought about it, though, I realized these
|
|
remain largely the same as they always have been: provide durable, scalable
|
|
storage for diverse use-cases without the need for specialized hardware and in
|
|
a way that allows both users and operators to be successful. In short, we will
|
|
continue to deliver the best object storage we can.
|
|
|
|
The exact details of how we do that may vary somewhat from cycle to cycle. The
|
|
last six months we've seen a lot of polishing of operational features, notably
|
|
container sharding and shrinking, part power increases, and error detection.
|
|
We'll likely continue working in these areas, and expand into larger
|
|
replication improvements.
|
|
|
|
I'll do what I can to help, both directly (by improving Swift) and indirectly
|
|
(by, for example, fixing broken gate jobs and flaky tests).
|