Todd, Answers inline:
On Jul 26, 2009, at 11:05 AM, Todd Greenwood wrote:
Flavio, thank you for the suggestion.
I have looked at the documention (relevant snippets pasted in
below), and looked at the presentations (http://wiki.apache.org/hadoop/ZooKeeper/ZooKeeperPresentations
),
but I still have some questions about WAN configuration:
---------------------------------------------------------------
WAN
----
A <-> B
A <-> C
A <-> D
A is a central processing hub (DC).
B-D are remote colo edge nodes (PODS).
Each POD contains (m) ZK Servers with (q) client connections.
---------------------------------------------------------------
What are the advantages and disadvantages to co-locating ZK Servers
across a WAN? Could you correct my admitedly naïve assumtions here?
1. ZK Servers within a POD would significantly improve read/write
performance within a given POD, v.s. clients within the POD opening
connections to the DC.
I'm assuming that you're setting the weight of ZooKeeper servers in
PODs to zero, which means that their votes when ordering updates do
not count.
If my assumption is correct, then you should see a significant
improvement in read performance. I would say that write performance
wouldn't be very different from clients in PODs opening a direct
connection to DC.
2. ZK Servers within a POD would provide local file transacted
storage of writes, obviating the need to write that code ourselves.
Yes, local zk servers in PODs receive all updates and process them as
any other zk server.
3. ZK Servers within the POD would be resilient to network
connectivity failure between the POD and the DC. Once connectivity
re-established, the ZK Servers in the POD would sync with the ZK
servers in the DC, and, from the perspective of a client within the
POD, everything just worked, and there was no network failure.
We want to have servers switching to read-only mode upon network
partitions, but this is a feature under development. We don't have
plans for implementing any model of eventual consistency that would
allow updates even when not being able to form a quorum, and I
personally believe that it would be a major change, with major
implications not only to the code base, but also to the semantics of
our API.
4. A WAN topology of co-located ZK servers in both the DC and (n)
PODs would not significantly degrade the performance of the
ensemble, provided large blobs of traffic were not being sent across
the network.
If the zk servers in the PODs are assigned weight zero, then I don't
see a reason for having lower performance in the scenario you
describe. If weights are greater than zero for zk servers in PODs,
then your performance might be affected, but there are ways of
assigning weights that do not require receiving votes from all co-
locations for progress.
-Flavio
