Pages: 1 2 3 4 5 [6] :: one page |
|
Author |
Thread Statistics | Show CCP posts - 23 post(s) |
Miilla
Hulkageddon Orphanage
60
|
Posted - 2011.09.13 15:23:00 -
[151] - Quote
Well it was a while back that dynamic node balancing was the ability to "hot" shift systems across nodes.
|
|
CCP Veritas
C C P C C P Alliance
104
|
Posted - 2011.09.13 15:35:00 -
[152] - Quote
Miilla wrote:Well it was a while back that dynamic node balancing was the ability to "hot" shift systems across nodes.
We call that "non-disruptive live remapping." It's something we absolutely want, but is still a ways away. There is an effort currently to bring disruptive live remapping - the ability to move a system between nodes after disconnecting everyone - up to par. Once that's done, we'll be prioritizing the work of making it non-disruptive against everything else we could be up to.
Time Dilation does help with that effort though - it provides a mechanism which we can use to pause the simulation of a system while it's moved, so we have some progress on that front without even trying CCP Veritas - Senior Programmer - EVE Software |
|
Miilla
Hulkageddon Orphanage
60
|
Posted - 2011.09.13 15:41:00 -
[153] - Quote
CCP Veritas wrote:Miilla wrote:Well it was a while back that dynamic node balancing was the ability to "hot" shift systems across nodes. We call that "non-disruptive live remapping." It's something we absolutely want, but is still a ways away. There is an effort currently to bring disruptive live remapping - the ability to move a system between nodes after disconnecting everyone - up to par. Once that's done, we'll be prioritizing the work of making it non-disruptive against everything else we could be up to. Time Dilation does help with that effort though - it provides a mechanism which we can use to pause the simulation of a system while it's moved, so we have some progress on that front without even trying
Frankly , the WINTER patch is CCP's last chance to stop the mass exodus.
You are running out of time.
|
Crasniya
Dragon's Legion of New Eden
4
|
Posted - 2011.09.13 16:06:00 -
[154] - Quote
Are "nodes" VMs running on the blades? Is there a software limitation which prevents you from moving VMs off a blade and giving that one node more of the blade's power, allowing it to handle a single overloaded node better?
Could one 'disruptively' move systems off an overloaded node that are currently devoid of players, so the interruption isn't noticed by anyone in-game? Without players in them, would moving about empty systems even help? |
|
CCP Veritas
C C P C C P Alliance
104
|
Posted - 2011.09.13 16:38:00 -
[155] - Quote
Crasniya wrote:Are "nodes" VMs running on the blades? Is there a software limitation which prevents you from moving VMs off a blade and giving that one node more of the blade's power, allowing it to handle a single overloaded node better?
They are not VMs - they're separate invocations of the server process. We are investigating ways in which we can use modern virtualization to solve the non-disruptive remap problem, but they're just investigations at this point as the problem is not trivial.
Crasniya wrote:Could one 'disruptively' move systems off an overloaded node that are currently devoid of players, so the interruption isn't noticed by anyone in-game? Without players in them, would moving about empty systems even help?
That's something we do occasionally and likely will do more once destructive remapping is a more reliable thing. Moving off empty systems doesn't really help anything at that moment, but when someone subsequently jumps into that previously-empty system, I'm sure they're happier having it been moved CCP Veritas - Senior Programmer - EVE Software |
|
Crasniya
Dragon's Legion of New Eden
4
|
Posted - 2011.09.13 18:13:00 -
[156] - Quote
Thanks for your answers. :) I'm sure EVE provides some... unique software requirements and limitations to keep it running smoothly, but I know with VMs we can move them between blades with effectively no interruption in service.
Keep up the good work. :D |
xxxTRUSTxxx
Galactic Rangers Galactic-Rangers
15
|
Posted - 2011.09.15 20:38:00 -
[157] - Quote
Crasniya wrote:Thanks for your answers. :) I'm sure EVE provides some... unique software requirements and limitations to keep it running smoothly, but I know with VMs we can move them between blades with effectively no interruption in service.
Keep up the good work. :D
agreed |
Josie Starshine
Republic Military School Minmatar Republic
12
|
Posted - 2011.09.15 21:15:00 -
[158] - Quote
Yes keep up the good work CCP. Some of us truly appreciate it. |
Sigras
Conglomo IMPERIAL LEGI0N
13
|
Posted - 2011.09.15 23:03:00 -
[159] - Quote
Miilla wrote:Well it was a while back that dynamic node balancing was the ability to "hot" shift systems across nodes.
Even if they implemented this, they've already said that the smallest chunk they can break eve down into is a one system chunk, so you'd just end up with one system on a node. This is already what happens when they reinforce a node.
What TiDi does is fix what happens in that system after that node which is only running one solar system gets overloaded.
Please make an effort to understand a system before you begin to criticize it. |
Vincent Athena
V.I.C.E.
59
|
Posted - 2011.09.15 23:19:00 -
[160] - Quote
Sigras wrote:Miilla wrote:Well it was a while back that dynamic node balancing was the ability to "hot" shift systems across nodes. Even if they implemented this, they've already said that the smallest chunk they can break eve down into is a one system chunk, so you'd just end up with one system on a node. This is already what happens when they reinforce a node. What TiDi does is fix what happens in that system after that node which is only running one solar system gets overloaded. Please make an effort to understand a system before you begin to criticize it.
I think there is more to it than that.
If a node running multiple solar systems gets overloaded by a big fight in one system it will go into TiDi, taking all the solar systems running on that node with it. TiDi is not limited to reinforced nodes. If at that point CCP could do dynamic load balancing they could move the big fight over to a reinforced node right in the middle of the fight.
All the people in the other solar systems are now happy because they are not in TiDi while they are just doing normal stuff.
All the people in the big fight are happier because the got a reinforced node all to themselves.
Now, if a fleet fight form was submitted and the solar system was already on a reinforced node, dynamic load balancing would not help. So it only helps for unplanned fights.
CCP employees should never proclaim a feature to be awesome. Only subscribers should. |
|
Triggadex
Black Dawn Merchant Industries
10
|
Posted - 2011.09.15 23:30:00 -
[161] - Quote
Personaly I think CCP are smarter then SOE, i mean SWG will be burnt into my brain for the rest of my life. CCP Guard's Day | Wake up --- Make Coffee --- Read and Reply on forums for 7 hrs --- Have meeting --- Sleep on couch in Office |
Largo Usagi
SregginWaffe Elite Space Guild
11
|
Posted - 2011.10.05 00:41:00 -
[162] - Quote
I just find it hilarious that CCP is fixing lag by giving lag a special name. Hey guys we are removing lag by making more lag, because that makes sense.
I'll reserve my full judgement of this feature when i see it in action but as it stands right now I see this as a round about way of solving the issue and hope that it will be a temporary feature; That in the future CCP will have worked lag out of the game where large fleet battles can be had without any slowing effect. I just see it as a handy cap for logistics pilots and that is about it. |
DarkAegix
Acetech Systems
66
|
Posted - 2011.10.05 01:41:00 -
[163] - Quote
Largo Usagi wrote:I just find it hilarious that CCP is fixing lag by giving lag a special name. Hey guys we are removing lag by making more lag, because that makes sense.
I'll reserve my full judgement of this feature when i see it in action but as it stands right now I see this as a round about way of solving the issue and hope that it will be a temporary feature; That in the future CCP will have worked lag out of the game where large fleet battles can be had without any slowing effect. I just see it as a handy cap for logistics pilots and that is about it. You clearly don't understand lag, time dilation, or anything else for that matter. I will speak simplistically in an attempt to help you understand.
Lag make gun shoot unfair. Weak man kill strong man. Make-server-slow kill lag. No weak man kill strong man. Make-server-slow only happen when lag. Server slow depend on lag. Very little slow if little lag.
If you can't understand that, then I'm afraid you haven't developed spoken language comprehension yet. Therefore, we will need to bang rocks and clubs against things, grunting and yelling, in order for you to understand. |
Pharago
Nughat Corp
1
|
Posted - 2011.10.05 02:48:00 -
[164] - Quote
CCP Veritas wrote:Crasniya wrote:Are "nodes" VMs running on the blades? Is there a software limitation which prevents you from moving VMs off a blade and giving that one node more of the blade's power, allowing it to handle a single overloaded node better? They are not VMs - they're separate invocations of the server process. We are investigating ways in which we can use modern virtualization to solve the non-disruptive remap problem, but they're just investigations at this point as the problem is not trivial. Crasniya wrote:Could one 'disruptively' move systems off an overloaded node that are currently devoid of players, so the interruption isn't noticed by anyone in-game? Without players in them, would moving about empty systems even help? That's something we do occasionally and likely will do more once destructive remapping is a more reliable thing. Moving off empty systems doesn't really help anything at that moment, but when someone subsequently jumps into that previously-empty system, I'm sure they're happier having it been moved
If a node was about to hit 100% TiDi, couldn't be possible to automatically look for a free reinforced node and copy the contents of the simulation to the new server process in the reinforced node, including a protocol to channel all current connections to that node?
That would allow for transparent reinforcement, as in changing to a new node without anyone noticing anything but a TiDi peak and then better performance.
Then, when all is over, the server could relinquish the reinforced node.
That would be a good one :)
|
Soldarius
Peek-A-Boo Bombers
8
|
Posted - 2011.10.05 03:28:00 -
[165] - Quote
Got-darnit, the forums ate my big-assed post! Again! :RAGE!::forum hate:
In short, great job gridlock. TiDi will not break the game. It will be an enormous improvement.
And plz remove supercap ewar immunity, kthxbai
PS, fix the damn forums from eating posts! ... plz "How do you kill that which has no life?" |
|
|
|
Pages: 1 2 3 4 5 [6] :: one page |
First page | Previous page | Next page | Last page |