PDA

View Full Version : Mega server!



xxHazexx
11-19-2012, 03:14 PM
If you haven't assumed this is about a server merge already im here to inform you. Yes this is a suggestion about consolidating all worlds into one. In my honest opinion this would be more efficient economically and help all communities with providing all players of all time zones more of a chance to run the content they've needed and don't feel like soloing with a hireling or multi boxing. Because this is an mmo and the majority of players that "generally" play mmo's intentions are to group with friends, guildies, or just enjoying a good PUG. Think about it: one game in particular that i know of "currently in development" is planning on using this and i think this would work exceptionally well the majority of the time anyway because everything is already instanced. As for public events well thats another suggestion with a solution one i haven't thought of yet. Just some food for thought.....Enjoy!

AlteredState
11-19-2012, 03:20 PM
Dunno, I kinda like having a unique name on my server. I don't want to be Character-1 Name or Character Name-1 or however they would "fix" that. Also imagine the potential nightmares that would occur with all the TR banks and shared banks.

*But if they were able to do that kind of thing smoothly I would be ok with it I suppose.

Syllph
11-19-2012, 03:25 PM
I would like to see different areas be servers instead of each world having a server. For example, the Marketplace is one server, Vale another.

Gkar
11-19-2012, 03:38 PM
In your expert opinion as a DBA and/or SysAdmin, how do you think the Ap server load balancing will be impacted by your suggested single server approach? Will they need to replace the load balancing tools from the ground up? How about the single DB/DB cluster, will it end up a system bottleneck?

As a customer service expert, what do you think will be the impact of the majority of players having one or more of their characters forcably renamed during the merger? How about the guilds names, some of which were originally paid for with real money/TP in the form of guild charters? What about the impact of reducing people from 2/4/10 characters per server to that number on a single server? Some people might be dozens of characters over limit. Please inform your CS answer by what happened the last time there was a server merge in DDO land and all the trouble that caused.

Memnir
11-19-2012, 03:40 PM
DDO would be hard pressed to emulate the upcoming Elder Scrolls MMO in regards to a Megaserver. Not the least of which being we have no idea of how Bethesda's concept will work in practicality beyond a ten second blurb in the preview video they put out last week or so.


However, Turbine is kinda stuck in terms of how they chose to implement names. It would irritate untold numbers of players if they had to rename their characters. Probably more so then it'd be worth. Likewise, it'd be a huge fuss and bother to get in there and recode the naming system at this point. For one - bugs. DDO Devs can't seem to make any changes to the game w/o causing ripple-effect bugs from popping up in unforeseen places. I don't even want to imagine what sort of things an attempt to revamp the naming system would screw up.


Plus - lag. For a fair percentage of the population, Special Events cause a lot of lag due to the raw number of instances open at the same time and what the players are doing in those instances. DDO's servers oft seem incapable of managing what is asked of them now. Compressing the entire population onto one server (even beefed up) would be nightmarish. And the cost of creating a brand new server, the architecture to support it, and then migrate all the data from the current servers to the new mega server is likely too prohibitive to even contemplate doing for an aging MMO like this.


/not signed

xxHazexx
11-19-2012, 03:46 PM
In your expert opinion as a DBA and/or SysAdmin, how do you think the Ap server load balancing will be impacted by your suggested single server approach? Will they need to replace the load balancing tools from the ground up? How about the single DB/DB cluster, will it end up a system bottleneck?

As a customer service expert, what do you think will be the impact of the majority of players having one or more of their characters forcably renamed during the merger? How about the guilds names, some of which were originally paid for with real money/TP in the form of guild charters?

Im no expert especially when it comes to server technology no worries though i plan on going to college to further my education some what in networking and similar aspects of the field. All that aside i am merely making a simple observation with simple logic. I pitch a logical rational idea and leave the "Experts" to the technical stuff.

xxHazexx
11-19-2012, 03:54 PM
DDO would be hard pressed to emulate the upcoming Elder Scrolls MMO in regards to a Megaserver. Not the least of which being we have no idea of how Bethesda's concept will work in practicality beyond a ten second blurb in the preview video they put out last week or so.


However, Turbine is kinda stuck in terms of how they chose to implement names. It would irritate untold numbers of players if they had to rename their characters. Probably more so then it'd be worth. Likewise, it'd be a huge fuss and bother to get in there and recode the naming system at this point. For one - bugs. DDO Devs can't seem to make any changes to the game w/o causing ripple-effect bugs from popping up in unforeseen places. I don't even want to imagine what sort of things an attempt to revamp the naming system would screw up.


Plus - lag. For a fair percentage of the population, Special Events cause a lot of lag due to the raw number of instances open at the same time and what the players are doing in those instances. DDO's servers oft seem incapable of managing what is asked of them now. Compressing the entire population onto one server (even beefed up) would be nightmarish. And the cost of creating a brand new server, the architecture to support it, and then migrate all the data from the current servers to the new mega server is likely too prohibitive to even contemplate doing for an aging MMO like this.


/not signed
I've quickly skimmed over your response with quite a bit of your information catching my eye. While all very rational i agree "to an extent". However you've only pointed out the cons in evaluating my suggestion. I know there would be issues with this but as days go by technology is changing if not "almost" daily. So again it's a given there would be some major issues with taking artifact technology and bringing it up too date or scrap it all start new with say...DDO 2? Import characters to one server and all that good jazz.

Gkar
11-19-2012, 03:56 PM
Im no expert especially when it comes to server technology no worries though i plan on going to college to further my education some what in networking and similar aspects of the field. All that aside i am merely making a simple observation with simple logic. I pitch a logical rational idea and leave the "Experts" to the technical stuff.

Fair enough.

As someone who has played with server load balancing inthe past, what I was trying to make clear is that this game, and many others, take a world/shard approach. We know very little about the technology "behind the curtain", but I'm sure the experts did that for some reason and that it wasn't a random "hey, lets not let everyone play together".

We also know that even with the existing server loads that during events (where user volume peaks) that the game often falls apart. This gives me zero confidence that putting 6x (or whatever) load on a single cluster would be a good thing, I'd bet the opposite. Also, while I have no idea what the bottleneck is, I think its probably safe to say that just adding all the CPUs together to create bigger app and server clusters probably would not make a more efficient one. If it was that simple I can't imagine that they would have paid the handful of dollars required to have their virtual servers and their server farm turned up a notch instead of messing around with fixes like stopping monsters from going "poof" when they die.

But hey, I'm just guessing too :)

xxHazexx
11-19-2012, 04:56 PM
+1 Brownie points to Sir Memnir for knowing what game i was referring too. I appreicate good observation. Attention to detail is hard to come by 😏

Memnir
11-19-2012, 06:03 PM
you've only pointed out the cons in evaluating my suggestion.Well, I for one don't see much point in merging the servers we have now into half their number - let alone merging all of them. I also think that the technological and practical-issue factors that bar the way are most likely insurmountable ones for a game that is as long in the tooth as DDO. I think this game has neither the hardware, solid enough coding, or nimble enough Devs to pull this off without catastrophic results.

Not to mention we really have no idea how Bethesda is going to implement the Megaserver idea in reality. From what they said about "the game figures out where to put you," - it seems far more likely that their game will have multiple instances of various zones within the game itself. I'm far from sold that Megaserver is anything more then a buzzword and ear-catching concept to drum up interest in MMO-jaded gamers. I'd really need to hear more about how they intend to pull it off before hitching my cart to that horse.
start new with say...DDO 2? Import characters to one server and all that good jazz.A DDO 2 wouldn't have a snowball's chance in hell of being based on D&D 3.5 rules, as the current DDO is -even if that foundation of 3.5 rules is slowly being eroded by the Devs. The upcoming Neverwinter is going to be based off the 4th edition, for instance. And WotC is at work on 5th edition rules even now. 3.5 is the past, and WotC tends to want new games to be reflective of their newest ruleset. So probably won't be mechanicly the same game, if it even gets made, which would make character transfers a super unlikely.



But, we all need our windmills to tilt at. No harm in your tilting at this one. I just think it's unlikely to be a giant, however.