![]() |
|
Register | FAQ | The Twelve Commandments | Members List | Calendar | Arcade | Find the Best VPN | Today's Posts | Search |
Gnotella (Windows) Gnotella has been discontinued. We highly recommend you use an actively developed client instead. |
| LinkBack | Thread Tools | Display Modes |
| |||
![]() #1 I actually have several ideas, but this one is the most important, for reasons that will becomes obvious as the idea is explained. If I connect to 4 gnotella servers, I should not only check their lists for who they are connected to, and cache them, but check the lists of the servers from that list and choose one of them to connect to (so that there are not redundant connections and wasted bandwidth on searches that have already gone through.) IE; I am system A. I connecto to system B, and find system C in his list. I momentarily connecto to C to get his list and choose one of those systems to connect to. Otherwise, I connect to system B and A and send redundant searches to both, only to have each of them send searches to each other, thus wasting bandwidth. This should be a user selectable setting for how far out my system should search for systems to connect to. Ideally this would be 4 or higher, so that a search sent on any 1 of the connections does not eventually wind up at any of the other systems I am connected to. I may not be explaining this clearly, but if you think about it, you'll understand why it's a great idea. (This should not, of course, apply to any of the caching servers) #2 using http is an upgrade to the gnutella/gnotella idea. PLEASE expand on this further. This will revolutionize gnotella if fully realized. #3 Search HITs should be cached to avoid multiple redundant searches, they should expire after 15 minutes or so. IE; System A sends a search to system B, who sends it to system C and the to D. D has the file and reports back through C and B and A that it does have it. The next time someone looks for a file that matches the filename returned from the first search, all four systems A,B,C and D will be able to report that system D has it. (system A should filter out any extra hits). #4 Turn on the web interface by default #5 somehow indicate which systems have the files and have the web interface enabled (this will save bandwidth, as users will be able to search directly) #6 discriminate against users who are NOT SHARING by lowering the priority on their searches, or lowering their 'queue priority', or how far their searches will go. |
| |
![]() | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Connection Lost or firewall is blocking connection error | rich31365 | Connection Problems | 11 | September 9th, 2007 08:56 AM |
connection problems, disconnects my entire internet connection | beels | Connection Problems | 3 | July 13th, 2007 05:56 PM |
UPC utilization when downloading | adnkladsnjklasnfd | Download/Upload Problems | 1 | August 1st, 2004 09:21 PM |
99% CPU Utilization - brings system to halt | Unregistered | User Experience | 6 | November 23rd, 2001 10:04 PM |
High CPU utilization | grem | Gnotella (Windows) | 20 | May 20th, 2001 12:18 PM |