Changeset 5045
- Timestamp:
- Sep 13, 2005, 2:46:58 PM (21 years ago)
- Location:
- trunk
- Files:
-
- 4 edited
-
Nebulous-Server/docs/design.txt (modified) (3 diffs)
-
Nebulous-Server/docs/requirements.txt (modified) (3 diffs)
-
Nebulous/docs/design.txt (modified) (3 diffs)
-
Nebulous/docs/requirements.txt (modified) (3 diffs)
Legend:
- Unmodified
- Added
- Removed
-
trunk/Nebulous-Server/docs/design.txt
r4943 r5045 1 1 # Copyright (C) 2004-2005 Joshua Hoblitt 2 2 # 3 # $Id: design.txt,v 1. 2 2005-09-02 04:20:29jhoblitt Exp $3 # $Id: design.txt,v 1.3 2005-09-14 00:46:58 jhoblitt Exp $ 4 4 5 5 Overview … … 23 23 technologies can be improved upon. 24 24 25 A centralized server model was used to allow efficient use of 'like' searches 26 on storage object names. The current 'best' technique for a distributed 27 metadata store is with distributed hash tables. Unfortunately, no widely 28 available DHT implementation allows 'like' searches on key names. 29 25 30 [1] Multiple Nebulous servers should be possible via database replication. 26 31 [2] A future version of Nebulous may include it's own data transport layer … … 31 36 32 37 The Nebulous Server and clients communicate via the SOAP protocol. Since SOAP 33 is platform independent this allows native Nebulous client to be written in a38 is platform independent this allows native Nebulous clients to be written in a 34 39 variety of languages. With in the Nebulous Server itself the IPC mechanism is 35 40 abstracted to allow ether multi-protocol support or complete replacement of -
trunk/Nebulous-Server/docs/requirements.txt
r2785 r5045 1 1 # Copyright (C) 2004 Joshua Hoblitt 2 2 # 3 # $Id: requirements.txt,v 1. 1.1.1 2004-12-22 02:16:23jhoblitt Exp $3 # $Id: requirements.txt,v 1.2 2005-09-14 00:46:01 jhoblitt Exp $ 4 4 5 5 - store all image data aquired during the first two years of PS1's … … 7 7 8 8 - injest new data from the summit at the same rate it is produced when 9 operating at a cadence of 40s between integrations. Approximately 400Mb/s. 9 operating at a cadence of 40s between integrations. Approximately 400Mb/s 10 and 64 files/integration. 10 11 11 12 - support phase 1 & 2 processing. x files per t_min … … 14 15 15 16 - support accessing small blocks of a file (ie., seek) 17 18 - tolerate partal system failures without data loss 19 20 - execeedly efficent use of both disk and network I/O 16 21 17 22 what are the science drivers for these 3? -
trunk/Nebulous/docs/design.txt
r4943 r5045 1 1 # Copyright (C) 2004-2005 Joshua Hoblitt 2 2 # 3 # $Id: design.txt,v 1. 2 2005-09-02 04:20:29jhoblitt Exp $3 # $Id: design.txt,v 1.3 2005-09-14 00:46:58 jhoblitt Exp $ 4 4 5 5 Overview … … 23 23 technologies can be improved upon. 24 24 25 A centralized server model was used to allow efficient use of 'like' searches 26 on storage object names. The current 'best' technique for a distributed 27 metadata store is with distributed hash tables. Unfortunately, no widely 28 available DHT implementation allows 'like' searches on key names. 29 25 30 [1] Multiple Nebulous servers should be possible via database replication. 26 31 [2] A future version of Nebulous may include it's own data transport layer … … 31 36 32 37 The Nebulous Server and clients communicate via the SOAP protocol. Since SOAP 33 is platform independent this allows native Nebulous client to be written in a38 is platform independent this allows native Nebulous clients to be written in a 34 39 variety of languages. With in the Nebulous Server itself the IPC mechanism is 35 40 abstracted to allow ether multi-protocol support or complete replacement of -
trunk/Nebulous/docs/requirements.txt
r2785 r5045 1 1 # Copyright (C) 2004 Joshua Hoblitt 2 2 # 3 # $Id: requirements.txt,v 1. 1.1.1 2004-12-22 02:16:23jhoblitt Exp $3 # $Id: requirements.txt,v 1.2 2005-09-14 00:46:01 jhoblitt Exp $ 4 4 5 5 - store all image data aquired during the first two years of PS1's … … 7 7 8 8 - injest new data from the summit at the same rate it is produced when 9 operating at a cadence of 40s between integrations. Approximately 400Mb/s. 9 operating at a cadence of 40s between integrations. Approximately 400Mb/s 10 and 64 files/integration. 10 11 11 12 - support phase 1 & 2 processing. x files per t_min … … 14 15 15 16 - support accessing small blocks of a file (ie., seek) 17 18 - tolerate partal system failures without data loss 19 20 - execeedly efficent use of both disk and network I/O 16 21 17 22 what are the science drivers for these 3?
Note:
See TracChangeset
for help on using the changeset viewer.
