About TODO Blog RSS Old blog Projects Gallery Notes

Thu, 15 May 2008

Meanwhile at appartment development side.

I installed vater system for the shower and thought to install the whole cabin, but found (as usual) that I do not have drills for the ceramic tiles. So, that will be postponed for a while.
Also I expect glue for ceramic tiles to be delivered today (as long as brick tiles), so that I can start hall granite covering. Although I'm a bit tired after water system installation, which took major part of the day.
It is actually simple task, but only when you have simple access to all parts. Now imagine a 10 sm thick wall, where you managed to drill two holes, each one about 2 sm in diameter (less than two fingers thick). In a meter below-left there is a bigger hole for sanitary (about 15x15 sm). Water system hatch is located 2.5 meters right to this.
Task is to put thin water tubes from water hatch to two small holes, but that splitter would be installed near bigger sanitary hole. Without direct access to any tube (you can only feel it, can not see) you have to connect them (also need to mention, that it is quite hard to put both hands into bigger hole for sanitary system) via different connectors using spanners.

I've completed the task, although not sure if it is really safe. That was challenging, and power sucking, so probably I will just slack this evening and hack some bits of captcha. Will also cover my table with the last colour level (yes, yes, it is still not done) and/or fill second varnish layer for x-shelves (they look really cool after mordant and varnish)...

/devel/flat :: Link / Comments (0)

POHMELFS distributed plans.

After healthy discussion started after my announcement of the second POHMELFS release, its time to highlight main ideas settled in the thread.

First, POHMELFS will be moved into parallel distributed filesystems, but still being very good as network filesystem. In particular, that will include ability to read data from one of the connected server (not particulary from currently active, how its done right now), writing will happen to all connected servers simultaneously (and transaction will be committed after all servers returned completion acknowledge).

Protocol will be extended to support dynamic addtion and removal of the servers to/from currently connected group. Probably there will be some kind of a status messages for servers (i.e. going offline, do not send me data, or I'm becoming slow, do not read from me and so on). It will be done in addition to cache coherency messages (I'm yet to implement, but because of other tasks, this was a bit postponed, probably to weekend), which will include two types of requests: page invalidation and inode update (that will also mean that POHMELFS will start supporting attributes (maybe even extended), right now it doesn't :). Such cache coherency protocol should scale better than classical MOSI (and its derivatives) and particulary better than pNFS spec proides (leases to operations for some servers), since it is still possible to work in parallel with the same file, especially without any overhead of data processing does not cross different client boundaries, but it has to be tested in practice.

POHMELFS server will be extended to support distributed facilities. Very likely it will be some kind of PAXOS algorithm, although probably in its very limited mode for the beginning. So far it will be really simple, so that I could touch all its corner cases and found optimal development strategy.

All client extensions are rather not that complex, although not always trivial, so that should not take too much time, so probably you will get something interesting soon.
Server extensions will be a bit slower, since I will start essentially from the distributed system ground and gradually move upstairs.

/devel/fs :: Link / Comments (0)