(Illustration by Gaich Muramatsu)
I am strongly thinking about using Coda as a solution to my current Ad-Hoc system of storing files on which ever computer I happen to be working on and sharing selected directory via SMB of NFS. But before I start I would like to know if Coda is up to the task. My Setup 100 MB Ethernet Network Linux/Windows 98 Desktop, 100 gig, PIII 500 Linux/Windows 98 Laptop, P 200 MMX Windows XP (Family), P4 Linux Server, 25 gig, P 166 MMX The Linux Server machine will be the main Coda server. If things work out it will get a larger hard drive. Desktop machine is mostly booted in Linux but occasionally in Windows. When in Linux it will act as a Coda Server and Client. When it Windows it will act as a Client. When used as a server its primary purpose will be for backup replication. The laptop will act purely as a coda client. If possible I would like the Linux and Win32 coda client to share the same cache. This is epically important on the laptop as disk space is tight (I only have 4 gigs) and because when I am disconnected I would like access to the same set of files in both Windows and Linux. The family computer will also act purely as a coda client. The main purpose use of this computer is for multimedia work. The primary reason for wanting to use CODA is that I do a lot of multimedia work which involves both my Desktop machine and the Family machine. Typically I will capture the the video on my Desktop machine. Then copy the files which vary anywhere from (100 - 2000 MB) over the network to my family machine and then process the files convert them to MPEG and move the final result back to my Desktop. I will then make an additional copy of the final results on the Linux Server for backup purposes. The files are transfered via SMB and NFS. I would like to use Coda so that I can avoid the manual copying of files around. It is infeasible to simply store the files on the Linux Server and use process them over the network since all file operations will involve network traffic which is simply not fast enough for multimedia purposes. I am hoping that Coda will cut down on network traffic significantly in order to make this feasible. In particular I am hoping that read and to a file will be almost as fast as accessing a local copy once the file is in the cache. I am also hoping that writing to a new file will be almost as fast as writing to the local hard drive so that I can capture directly to the coda server. I assume code were store the file locally in the cache and will write the changes back to the server when the file is closed to provide high write throughput. Even if store everything on the Linux Server I still have to worry about making manual backup copies. Using my Desktop machine as a coda server for backup purposes will solve this problem nicely. If coda works out I will also be storing a significant amount of other files (non-multimedia) currently on my Desktop on the Linux server so that I can seemly access the files on my Notebook without a lot of fuss. Currently I have to remember to copy the files I need access to on my notebook. Simply sharing the files via NFS or SMB on my desktop won't work because often I will be using Windows on my Desktop machine and will need access to files on my Linux partitions from my Notebook. So, I would like to know if CODA is up to the task? In particular I would like to know how stable coda is on Win32 and how well it will deal with the huge multimedia files. I would also like to know if my Linux Server is powerful enough to act as a CODA server. Consistent read and write throughput is also important, especially when capturing video. I expect read throughput to be consistent for reading large files which are not yet in the cache so the playback will be smooth. I hope I explained everything sufficiently. If not let me know and I will try to clarify. -- http://kevin.atkinson.dhs.orgReceived on 2002-10-15 13:29:44