Jump to content
LOTROCommunity

A random sampling of Server populations. Is it time for Server merges?


Recommended Posts

It's not just calculating how much it costs to merge the server that's relevant, one also has to calculate long term administrator/network savings as well.

 

What would your racks look like with 1/2 the current servers? How much would you save from reduced DataCenter costs?

How many Admin hours do you save by reducing the number of servers that need to be kept updated and backed up? Enough to lay off one or two Admins?

Wouldn't a server merge be a good time to prune F2P accounts that haven't had activity in say 2 years? Reducing the size of your storage always has cost benefit.

 

While there would be considerable pain in server merges, I honestly think the savings would balance the long term scales of the debate...

It think most here agree that world are split between computing instances and that computing instances are already put together on single piece of hardware.

So the number of physical servers would be more dependent on total game population and only go down a little by putting the same number of people on fewer world names.

Link to comment
Share on other sites

It think most here agree that world are split between computing instances and that computing instances are already put together on single piece of hardware.

So the number of physical servers would be more dependent on total game population and only go down a little by putting the same number of people on fewer world names.

I find it next to impossible that this entire game runs on a single piece of hardware.

 

the instances are likely virtual and split up by virtue of the resources they need to run, one host machine for each number of instances that host is equipped to handle. Could the host machines under their current load assume another instance or two? I don't know but I bet they could and consolidating the instances could free up entire host machines.There's also all of the supporting  applications and appurtenances as well such as the load balancers, firewalls, switches application servers and the rest of their infrastructure in their network.All of these devices contribute to their DC footprint and reducing that footprint through consolidation would save money.

 

I think the largest costs savings would come in manpower hours though, that's always the largest cost in any operation like this. Consolidation could potentially save them the salary of an Admin or two and I bet that would be fairly significant.

Link to comment
Share on other sites

Who said single?

 

You did? :P

 

 

It think most here agree that world are split between computing instances and that computing instances are already put together on single piece of hardware.

So the number of physical servers would be more dependent on total game population and only go down a little by putting the same number of people on fewer world names.

Link to comment
Share on other sites

Saturday, May 31, 11:30 PM  P.S.T.

 

Meneldor:

GLFF: 52

Global 122

 

edited to add:

 

11:35 PM P.S.T.

 

Gilrain:

GLFF: 13

Global: 30

Meneldor  sad.

I remember only 1 year ago seeing the glff numbers in the mid 200's.  Now you can't even scrape 5 raids together.  Our kinalliance used to run 3 raids at a time.  This is also the troll filled GLFF, if you think your home servers GLFF is bad come visit Meneldor, where you can easily find out a persons life story by reading for 3 minutes.  You can also hear my buddy Kickman talk about how healthy the game is there.  Also the come at me brah twins.  Meneldor GLFF otherwise known as a troll cave. 

 

The Gilrain numbers are a joke.  Any developer that lets a server population that low not freely transfer is doing that server injustice.  Even if "not everyone used the GLFF" that is pathetic. 

 

One wonders how the hobbit to isengard went there?  Musta been embarrassing.

Link to comment
Share on other sites

The Gilrain numbers are a joke.  Any developer that lets a server population that low not freely transfer is doing that server injustice.  Even if "not everyone used the GLFF" that is pathetic.

I would not read to much into the Gilrain figures posted Gilrain is one of the former EU srevers and the figures were taken at 23:35 PST, which is very early in the morning in europe

Link to comment
Share on other sites

I would not read to much into the Gilrain figures posted Gilrain is one of the former EU srevers and the figures were taken at 23:35 PST, which is very early in the morning in europe

 

Sunday, June 1, 9 AM p.S.T.

 

Gilrain:

GLFF: 50

Global: 114

 

i will revisit certain servers at random  (my whim)  and i have conceded that some times will be from peak hours & some will be from off peak.  these are random samplings, and for any big picture, one needs a good sample size.

 

 

Really, the only thing that matters is how much it would cost for Turbine to merge servers (a lot), and is that likely to earn them any more revenue (no, they'd likely lose customers) or reduce expenditures (maybe an extremely tiny bit for operations and maintenance).

 

I think there may also be some corporate prohibition against merging servers at Turbine.

 

Although Microsoft merged some of AC2's servers when they were still under its control (and sold it off a couple months later), I don't think Turbine has ever merged servers in any of its games. Even though the original Asheron's Call got its last ever update in March, it still has the same 9 (mostly empty) servers it's had since 2004.

 

 

the cost to merge must be weighed against the cost of maintaining a server that has fallen below a threshold of players.  at a certain point, the cost of doing nothing will factor in.  i argue that on some servers, this cost has started.

 

again, if you do not intend to keep the game online long term.  of course you would never absorb this cost of merges.

 

as to has Turbine ever merged servers?  Yes, yes it has.

 

 

 

Server merge announcement from Turbine, in 2007

We have some very important news for our players regarding Dungeons and Dragons Online™: Stormreach™ (DDO).

Dungeons & Dragons® has always been about having a good time with a party of your friends. With DDO, we bring that experience to life online for the first time, and appreciate all the time you, the players, spend inside Stormreach fulfilling the promise of D&D®.

Over the last few months we have seen an increase in the number of players asking us to make it easier to find groups. Since grouping is such a core part of the DDO experience we looked at a variety of ways to make the grouping process easier for players of all levels. Various options were considered, but the best way to make a change that improves the situation for everyone is to concentrate active players on a smaller number of servers.

To this end we have developed a plan to merge the game’s servers which will greatly improve player experience across the board while also improving the performance of the game by upgrading to a stronger infrastructure on the back end.

We want to thank you for all of your patience as we implement these changes to the game.

 

[edit] New Server Details

We'll be combining the source servers on the left into the destination servers on the right.

Before we get started with the complete server merge, we are planning to run a public preview on Risia. On Wednesday, July 25th, we will be bringing Risia up with the Mabar/Argonnessen server merge and we'd love it if you'd hop onto Risia and help us confirm everything is operating as expected. Any players that have created a character on Mabar or Argonnessen before July 23rd are invited to participate in the merge preview.

Once we have completed the merge preview and evaluated the results we expect to begin merging the main servers starting the following week.

The process of merging the servers will be divided into two downtimes.

  • Khyber, Aundair, Riedra, Thelanis, Xoriat, and Tharashk will be offline on Thursday, August 2nd, between 2am and 7pm EDT. Around 7pm, Khyber and Thelanis will come back online and contain all of the merged characters and data from their respective servers.

  • On Monday, August 6th, Fernia, Ghallanda, Argonnessen, Lhazaar, Adar, Sarlona, Aerenal, and Mabar will be offline from 2am until approximately 7pm EDT. At about 7pm, Argonnessen, Ghallanda and Sarlona will come back online and will contain all of the merged characters and data from their respective servers.

When all service resumes on August 6th, the DDO servers will be: Argonnessen, Ghallanda, Sarlona, Khyber, and Thelanis.

On August 13th, all servers will be offline from approximately 7am until 1pm EDT, for server maintenance and optimization after the merge.

 

 

http://ddowiki.com/page/Server_merge

Link to comment
Share on other sites

It's not just calculating how much it costs to merge the server that's relevant, one also has to calculate long term administrator/network savings as well.

What would your racks look like with 1/2 the current servers? How much would you save from reduced DataCenter costs?

How many Admin hours do you save by reducing the number of servers that need to be kept updated and backed up? Enough to lay off one or two Admins?

Wouldn't a server merge be a good time to prune F2P accounts that haven't had activity in say 2 years? Reducing the size of your storage always has cost benefit.

While there would be considerable pain in server merges, I honestly think the savings would balance the long term scales of the debate...

Just some random thoughts:

The key variable in your scenario is "long-term." How long-term? A year? Two?

Backing up should be easily automated, so that doesn't even count. Updating? There won't be much to update at all for the foreseeable future. If things are running stable now with minimum interference (no systemic infrastructure changes, only minor geography additions and some daily quests, no raids, reused textures, etc.), couldn't just one admin handle all that already? It's not like they have to take into account loads at this point, and there's no need to heavily BR test anything either. They could just cut out BR entirely, avoid the whole test setup.

As for storage space, that's the cheapest variable out of anything these days.

I kind of suspect pride may be a hidden variable in this: Even considering server merges would be OPEN ADMISSION that LotRO is failing---none of the official stances from Rick Heaton upwards have given that indication.

Link to comment
Share on other sites

I kind of suspect pride may be a hidden variable in this: Even considering server merges would be OPEN ADMISSION that LotRO is failing---none of the official stances from Rick Heaton upwards have given that indication.

 

bloody christ on a cracker this mentality is soo...  retro.  old people need to stop flapping their lips with it... already!

 

World of Warcraft addressed it's dwindling populations with cross server grouping.  Blizzard will likely re-dress as time passes, because they don't tend to drop the ball...  which is why they are (still) the MMO industry behemoth.

 

GW2 has taken it's existing overflow system and rejiggered it to function like a mega-server, thereby addressing the issue.

 

ESO launched with mega-server infrastructure.

 

whilst SONY my have had to swallow pride, they didn't wait till it was too late to merge SWTOR servers.  & the community there was vocally asking that population issues be addressed.

 

a seven year old game merging servers isn't failure.  it isn't an admission of anything, other than addressing a games aging population and making the structure that population is housed in, fit.

 

if this was still 2008,  the expression of merges is a sign of failure may still hold water, but it is 2014...  and in accelerated internet time, that is like a ten thousand year old dog.

 

the only thing i see it as admission of is...  we will pull this game offline, in a heartbeat.

 

games that intend to continue, address population issues.

Link to comment
Share on other sites

On probably any other gaming OF, threads about "this game is dying" and low populations and whatnot are allowed to stay (until things degenerate, of course); Turbine actively closes and deletes such posts, though. So yeah, Turbine is "special." Yes, I'd expect server merges would be a big admission on the official LotRO forums after all the unhappy players have been so openly chased off by the remaining people with full Sapience blessings, and I do not believe the remaining population has been heavily asking for server merges. Keep in mind that they would have to basically admit that Helm's Deep was a failure that led to this current situation in the first place---I certainly don't see that happening. It took years for whatshisface of SOE to admit how CU/NGE was a clusterfuck (and what does Sony have to do with SWTOR?).

Your ideas make sense for a typical MMO seeking to succeed in the genre under rational leadership. Neither apply to LotRO.

Link to comment
Share on other sites

Working on the assumption that Turbine could or would not merge worlds as Blizzard did with WoW, so allowing the merging servers to keep their old communities, guilds and character names, there would be absolute uproar if Turbine forced people into the lottery of deciding which character or kin name would be the one to remain as is post server merge. Just that alone will stop Turbine even considering a merge and will just let things run as they are.

 

In any event, it's by and large a solo game now anyway so whether there are 10 or 10,000 other players online is largely irrelevant.

Link to comment
Share on other sites

 

 

as to has Turbine ever merged servers?  Yes, yes it has.

 

 

 

This is fascinating, and it SOUNDS like Turbine did the DDO server merges for all the right reasons, too. I take back what I said regarding a corporate culture prohibiting server merges.

 

Turbine has always said LOTRO doesn't need to merge servers because populations are healthy. It makes me wonder what Turbine would consider unhealthy.

Link to comment
Share on other sites

and i will borrow that old canard Nymphonic so enjoys touting about when some of these arguments are made...

 

"that old game Ultima Online is still around after all these years."

 

yes it is around.  how many servers does it have?   one.  and with the help of server emulation, that one server hosts the current player base.

 

and that is the point i am making.  "if" Turbine intends to continue to host this game, as opposed to shutting it down cold, out of the blue.  at some point this consolidation will have to take place.

 

that, or they pull the game offline cold.

 

Can they follow the Ultima route, given IP situation? I assume they have to pay IP fees regardless of how many servers they have, as long as they are running the game. So pulling it offline seems like the route they would have to take.

Link to comment
Share on other sites

Can they follow the Ultima route, given IP situation? I assume they have to pay IP fees regardless of how many servers they have, as long as they are running the game. So pulling it offline seems like the route they would have to take.

 

No one really knows why kind of licensing fees and royalties Turbine pays to SZC. However, given SZC's cozy licensing arrangement with WB in general, it seems unlikely SZC would demand the kind of fees that would drive the game into unprofitability and shutdown, at least as long as there's no other potential MMO licensee on the horizon.

Link to comment
Share on other sites

Who said single?

 

 

"computing instances are already put together on single piece of hardware."

 

 

 

My apologies if you meant multiple pieces of hardware, I read that as a single piece piece of hardware

 

Since you agree there are multiple pieces of hardware, you might also agree that my point has merit and merging servers could reap some cost benefit.

Link to comment
Share on other sites

My apologies if you meant multiple pieces of hardware, I read that as a single piece piece of hardware

 

Since you agree there are multiple pieces of hardware, you might also agree that my point has merit and merging servers could reap some cost benefit.

Some, but not linear to the reduction of worlds.

If you now need 50 pieces of hardware to hold 20 worlds for computing load reason then reducing it to 10 world will probably save 8, not 25, computers. Assuming same number of players.

Of course an enforced merge will see an extra player drop. Every downtime LOTRO has now sees people drop out permanently, that is very visible in the charts. Plus some will come back and get pissed about details of the result such as renamings. Some more will then meet rude people they don't want to play with that were previously on a different server. Some more will leaves over temporary screwups that need tickets to resolve such as shared account items having hickups (and there will be plenty).

Link to comment
Share on other sites

I'm not sure how/if it translates to LotrO hardware, but WoW Private servers can be handled on very reasonable hardware according to forums like this:

http://www.ownedcore.com/forums/world-of-warcraft/world-of-warcraft-emulator-servers/301946-dedicated-server-specs-needed-run-wow.html

http://www.ac-web.org/forums/archive/index.php/t-127225.html

Link to comment
Share on other sites

Not that I put much stock in that hype propaganda, but for what it's worth, an item that came up in recent hobbit run:

 

Has Turbine ever thought of server merging to cut down on the amount of servers
Rowan – From his perspective as EP. They launched the game in 2007, and technology has come a heck of a long way. They did a major technology upgrade around 2012 when they went to F2P, and powered up a lot of their server infrastructure, but things keep increasing, and increasing dramatically. So he thinks something like that could potentially leverage change for them, potentially help make that an even better experience, and help get more people playing together.

 

https://www.lotro.com/forums/showthread.php?539143-Hobbits-to-Isengard-Q-A-with-Sapience-notes-on-what-was-said&p=7175808#post7175808

Link to comment
Share on other sites

whilst SONY my have had to swallow pride, they didn't wait till it was too late to merge SWTOR servers.  & the community there was vocally asking that population issues be addressed.

SONY has absolutely nothing whatsoever to do with SWTOR, that game was designed, is owned and is run by EA/Mythic/Bioware

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...