Page 1 of 2 12 LastLast
Results 1 to 40 of 74
  1. #1
    Join Date
    Jul 2006
    Posts
    74

    Mediatemple's Grid Server, is it safe yet?

    I have been eyeing Mediatemple's grid server for quite some time, and have followed up on it heavily on these forums. I have noticed that the MT posts have lost steam, and I am wondering if this is because they have straightened out the problems, or just because people have gotten tired of their problems, and switched back? Is it safe for me to open an account on the gridservr? From what I have heard, the main problems are the ones listed below. Have they been resolved, or are some of them still affecting sites on the grid?
    1. The 403 errors encountered when visiting sites on the grid
    2. MYSQL connection errors
    3. Account center slow
    4. blatant downtime
    Also, I hear that you have to get a SSL cert to get a dedicated IP. Is that really true, and if so, what would not having a dedicated IP mean for my site other than harder DNS management?
    Thanks,
    Ben
      0 Not allowed!

  2. #2
    Join Date
    Oct 2006
    Location
    canada
    Posts
    527
    1. I've personally never had any of the 403 errors on either my test sites or on any of the more active sites that I have on the grid.

    2. No errors, but it is still kinda slow to respond.

    3. It's gotten a lot faster then what it was before, at this point it's probably faster for most things then cpanel was on my previous host. However that's super subjective as it's different everywhere.

    4. No idea as I don't actively track uptime, probably should =/
      0 Not allowed!

  3. #3
    Join Date
    Jul 2006
    Posts
    74

    Thanks

    Thanks, It's good to hear they are actually making some progress..... Do you use rails? my dev is centered around rails, and it would be good to hear from some rails users. Thanks,
    Ben
      0 Not allowed!

  4. #4
    Join Date
    Dec 2003
    Posts
    543
    A buddy of mine uses MT's grid server and I haven't seen any issues with his sites hosted there.
      0 Not allowed!

  5. #5
    Since there launch i have expirience 3hr and 30 minutes downtime. This month so far only 15 minutes.

    Speeds have gotten better but of course I use it for nothing production, the mysql errors look to be getting better.

    If MediaTemple can pull this off then it could change the hosting industry drastically.
      0 Not allowed!

  6. #6
    Join Date
    Jul 2006
    Posts
    74

    If they can pull it off

    If MediaTemple can pull this off then it could change the hosting industry drastically.
    Good point, that's the only reason I am still investigating this despite the problems... But they absolutely HAVE to get their act together, because they have all this PR about the grid server, and if they pulled it because of too many problems, think of the backlash. However, I am still bummed out about having to pay 50$ + 10$ a month for a dedicated IP .
      0 Not allowed!

  7. #7
    Join Date
    Jun 2004
    Location
    Bay Area
    Posts
    1,320
    This Grid solutions seems very interesting actually. However it still is shared hosting. What kind of performance would I be able to get on this? Would this, because its on a grid, be able to handle a site with 300,000 dynamic PHP hits for example?

    I cant find any prices for "CPU over usage" on their site actually...does anybody happen to know what they charge?
      0 Not allowed!

  8. #8
    Join Date
    Jul 2006
    Posts
    74

    Don't quote me on this

    Don't quote me on this, but I think it is 10 cents a GPU, but that's not official....
    Although I have also heard they charge a hefty 2.50 for extra Bandwith.
    Thanks,
      0 Not allowed!

  9. #9
    did not read a lot about this grid, but I do not like to put all eggs in same basket, or even put all servers in same data center.

    some times server got hacked and all indexes replaced by hacker message, what will happend to the gird in such cases? .. may be a naieve example .. but this happend last month for hostgator and I wonder what will be the case if hostgator was using a grid tech.
      0 Not allowed!

  10. #10
    Join Date
    Mar 2006
    Location
    New York USA
    Posts
    402
    Ok guys, i keep hearing this whole "grid server" thing and raves about it. From what my associates and i grasp about the grid server, is that simply it's just a fancy name for a cluster of servers, as well as a RAID array, which all provide fault tolerance. RAID levels vary per your needs, but mostly focus on speed of data storage/requests, and fault tolerance.

    With raid levels that allow for disk mirroring, this basically keeps an exact copy of Drive A on Drive B, so that if Drive A Fails, drive B can now become Drive A with the exact information and data that the Original Drive A had, and causes only a few seconds of downtime.
      0 Not allowed!

  11. #11
    Join Date
    Jul 2006
    Posts
    74

    Cluster, Not Grid

    it's just a fancy name for a cluster of servers
    I would have to agree; I remember before it came out, and when it was still SS 6.0, the preview page called it (mt) cluster technology, so "GRiD" is definetly just marketing. Even if MT's GS doesn't work out though, their marketing hype will bring attention to some of the other providers offering grid hosting such as thegridlayer, mosso, RailsMachine etc. The only problem is that MT's is the only managed one, but also the only one with major glitches. The grid layer has you configure application servers, firewalls, and mysql servers, so it may be out of the question. As for some others, their just too pricey.
    Last edited by marak; 11-12-2006 at 05:05 PM.
      0 Not allowed!

  12. #12
    it's just a fancy name for a cluster of servers
    I would have to agree
    Agreed, this isnt a grid at all... though, it is a nice HA cluster formation...

    Basically, they are using load balancing technology across multiple servers with a redundant storage array in the backend (bluearc or netapp or something similar)

    Having said this - there is no technology that can accomodate 1000 GB of transfer for $20/month - this is a marketing gimmick -

    Would this, because its on a grid, be able to handle a site with 300,000 dynamic PHP hits for example?
    No way, no how - and if they somehow could accomodate this, the fees for CPU overages would be huge

    1000 GB for $20 is overselling the overall CPU capability within the "grid" (ie nodes) - and with the oversold affect, each account actually has access to less CPU then in an environment that is not oversold. So, the overages you would need to pay for using 100 or 200 GB of heavy DB traffic due to CPU overages would be pretty steep - they would have to be in order for the model to work...

    There are many hosts with clustered environments that have better uptime and performance records then MT or Mosso (not too mention a greater feature set) - and the total cost of ownership for the average user or heavy user is much lower. Additionally, more and more hosts are implementing this HA "grid" type of solutions with LB nodes and shared, redundant storage arrays - go with one of these companies that have reasonable price points - this 1000 GB transfer for $10-$100 is a gimmick at best and will not save you money in the long run...
      0 Not allowed!

  13. #13
    Join Date
    Jul 2001
    Location
    Indianapolis, IN
    Posts
    537
    Quote Originally Posted by CartikaHosting

    There are many hosts with clustered environments that have better uptime and performance records then MT or Mosso (not too mention a greater feature set) - and the total cost of ownership for the average user or heavy user is much lower. Additionally, more and more hosts are implementing this HA "grid" type of solutions with LB nodes and shared, redundant storage arrays - go with one of these companies that have reasonable price points - this 1000 GB transfer for $10-$100 is a gimmick at best and will not save you money in the long run...
    Can you give any examples? Not arguing, just curious
    neil MCITP, VCP
      0 Not allowed!

  14. #14
    Can you give any examples? Not arguing, just curious
    Examples of clustered hosts with better uptime then MT/Mosso?

    Sure - JodoHost, VortechHosting, DynamicNet (pretty much any well run H-Sphere provider that has a clustered setup)

    Not too mention the inherent dual platform support and full featured control panel - real reseller features, etc.....

    As for hosts implementing LB - I will need to look around - I have stumbled across a few over the years - and I assure you, alot more are on the way.... the important thing to note is that not all of them (actualy most of them) - are not trying to sell 1000 GB for $20
      0 Not allowed!

  15. #15
    Join Date
    Oct 2006
    Posts
    51

    Media Temple Grid Server Issues

    Quote Originally Posted by marak
    I have been eyeing Mediatemple's grid server for quite some time, and have followed up on it heavily on these forums. I have noticed that the MT posts have lost steam, and I am wondering if this is because they have straightened out the problems, or just because people have gotten tired of their problems, and switched back? Is it safe for me to open an account on the gridservr? From what I have heard, the main problems are the ones listed below. Have they been resolved, or are some of them still affecting sites on the grid?
    1. The 403 errors encountered when visiting sites on the grid
    2. MYSQL connection errors
    3. Account center slow
    4. blatant downtime
    Also, I hear that you have to get a SSL cert to get a dedicated IP. Is that really true, and if so, what would not having a dedicated IP mean for my site other than harder DNS management?
    Thanks,
    Ben
    To answer your questions:

    1) No problems with 403 errors
    2) Flagrantly bad (is that a term). My sites were all inaccessible on Saturday. I reported the trouble. It took 6 hours to get a response.

    This is their response:

    Support response: 2006/11/11 13:14 We have looked into the issue that you have reported and unfortunately could not find any problems with your service. Furthermore, using a series of diagnostic checks we were not able to reproduce any of the problems that you reported. We have been monitoring your services and all features are operational.

    It is possible that either the issue has resolved itself through standard account resetting or you have taken actions to resolve the matter yourself. We have tested your hosting services and all its components are working to specification.

    Please continue to update us, using this inquiry number, if you experience and further difficulties regarding this subject.

    Of course how am I to know that it was MYSQL? I don't. But it sure seemed like it.

    3) Account Centre - No problems. But if they are running it on their own Grid Servers and use MYSQL - well figure it out.

    4) Blatant downtime - Ha! You bet. See 2 above

    To cap it all off tech support takes forever to respond. And when they do, the answer is useless (canned). I think these poor guys are swamped. I am working on another (unbelievable) MYSQL import issue. I haven't posted the string on this Forum yet as it is getting more and more intertesting everyday. Once it gets resolved, I'll post the whole thing.

    I hope this answers some of your questions.

    I would hold off on using MT as they clearly do not have their act together yet.

      0 Not allowed!

  16. #16
    Join Date
    Aug 2005
    Location
    Canada
    Posts
    838
    Quote Originally Posted by CartikaHosting
    Examples of clustered hosts with better uptime then MT/Mosso?

    Sure - JodoHost, VortechHosting, DynamicNet (pretty much any well run H-Sphere provider that has a clustered setup)
    I guess many people talk about "clustered host", thinking as "clustered hosts with some degree of server redundancy".

    Simple H-Sphere cluster doesn't saistfy them unless at least web servers are redundant, IMO.
    (Although I think Clusterd setup without redundancy, like DreamHost and many other hotsts, are better than single server setup.)

    I guess having RAID storage is becoming "standard" or "normal",
    and if a host doesn't have RAID or other form of redundant storage,
    it can be a factor for avoiding the host.

    As for hosts implementing LB - I will need to look around - I have stumbled across a few over the years - and I assure you, alot more are on the way.... the important thing to note is that not all of them (actualy most of them) - are not trying to sell 1000 GB for $20
    Even low budgets hosts are using Loadbalancers. Servage, for examle.
    Even somewhat infamous ****** seems to be using.

    iWeb offers loadbalanced shared (reseller) hosting, too.
    They have hot standby MySQL server, as well.

    And LayeredTech has started Grid server (for those who want more control than MT or Mosso can offer).

    I guess the list will continue to grow, and the price will drop.
    Good for users, but possibly challenging for some hosts.
      0 Not allowed!

  17. I purchased the grid hosting not too long ago after they ironed out the major problems.

    1. The 403 errors encountered when visiting sites on the grid

    None.

    2. MYSQL connection errors

    None. I currently only have a 'test' forum up at http://wiiglobal.com/forums/ and it has been loading very fast with no errors. My site will launch on Saturday with a very complex CMS which will also be using php and mySQL. I'll let you know how things go.

    3. Account center slow

    Fluctuates from slightly slow to moderately fast. So not really a serious problem.

    4. blatant downtime

    None.
      0 Not allowed!

  18. #18
    Join Date
    Oct 2002
    Location
    Behind You
    Posts
    791
    Yeah, i trust the new guy who has only posted positive about MT and who who has admitted to purchasing before these reported problems... I guess they gotta try something to stop the flow of bad press from real users.
      0 Not allowed!

  19. #19
    Join Date
    Oct 2006
    Posts
    51
    Quote Originally Posted by amplifiedshock
    I purchased the grid hosting not too long ago after they ironed out the major problems.
    I guess you are not on GS666! I am - enough said.

    Les
      0 Not allowed!

  20. #20
    Join Date
    Jul 2006
    Posts
    74

    New accounts

    I guess you are not on GS666! I am - enough said.

    Les
    Wait - so new accounts wouldn't have this problem, but old ones would?
      0 Not allowed!

  21. #21
    Join Date
    Oct 2006
    Posts
    51

    * Whoa is Me with MT GS!

    Quote Originally Posted by marak
    Wait - so new accounts wouldn't have this problem, but old ones would?
    I was being sarcastic. If someone is claiming to have no problems on MT's GS offer while Media Temple is acknowledging that they are still having MYSQL problems, then that person is:

    1. not running sites with MYSQL
    2. not monitoring their sites effectively
    3. not telling the truth (oops - faux pas)
    4. has a vested interest in MT and is trying to put a positive spin on things
    As of this writing my sites are all beginning to slow down as they have every day since I moved to MT. BY 3 PM EST they will be either super slow or not responding.

    Any suggestion that I am "bashing" MT because I hate them or because I am trying to sell my own (or those of others) services are ill-placed. I simply want to get these issues resolved and enjoy the high QOS MT has promised.

    Les
      0 Not allowed!

  22. #22
    Join Date
    Jul 2006
    Posts
    74

    Mt

    Any suggestion that I am "bashing" MT because I hate them or because I am trying to sell my own (or those of others) services are ill-placed. I simply want to get these issues resolved and enjoy the high QOS MT has promised.
    You're definetly not bashing them. They unleashed their PR monster, making arrogant comments like "The grid won't go offline until California goes into the ocean", and all these other things. After that kind of PR, and all their 101% uptime and gimmicks, they should provide what they say. I think we should continue to comment until they get their act together.
      0 Not allowed!

  23. #23
    Join Date
    Oct 2002
    Location
    EU - east side
    Posts
    21,913
    As of this writing my sites are all beginning to slow down as they have every day since I moved to MT. BY 3 PM EST they will be either super slow or not responding.
    That resembles overloading during peak times. It could be that the success of their marketing department really exceeded all their expectations.
      0 Not allowed!

  24. #24
    Join Date
    Jan 2003
    Location
    Texas, where else?
    Posts
    1,571

    Cool

    As I have said in other posts the "Grid" concept is interesting but doesn't seem to have the bugs out.
    Then you have load balanced clusters etc.
    However I keep being reminded of an interesting article I read a long time ago (before all the "grid" stuff) about as a host grows whether to go with huge servers & bigger RAID arrays or clustering or whatever. The jest of the article (and I understand that conceptually the "grids" are supposed to eliminate this but obviously it's not all smooth sailing yet) was that with big servers or clusters etc. you may still have single points of failure (even if it's just a switch) The author's "food for thought" was "is it better to have a monster system that if it fails or slows to a crawl results in tons of mad clients, or is it better to have a larger quantity of more "standard" servers so if 1 fails or slows down you only have the support issues from the people on that one box"? Things like H-sphere and newer load balancing & clustering set-ups (and more advanced versions like the grids) may be changing that perspective some, but I can't help but still wonder about that basic idea.

    I'm sure these folks have teams of admins pulling their hair out trying to get everything running smoothly. But these are still just hosts. Recently I read another article (I guess I like to read about other people's problems ) where an entire floor of servers in a new state-of-the-art super-redundant datacenter was brought down by a power failure that "rolled" through all the redundant systems. Their on-site highly skilled techs found the problem and used "jumpers" to get things on-line then a high-priced failure forensics team came in to find out what the heck really happened. It all came down to a little electrical connector in one circuit box, not even a $1 part. This led to analysis of the entire center where they found 20 some of these little connectors in various locations had the same corrosion problem and could potentially fail. This led to thousands of dollars in replacing all these little connectors in all the boxes in the facility. Evidently a "bad batch" of connectors the electrical contractor was supplied. My point being if this new-state-of-the-art center could be brought down by one little connector resulting in the malfunction of all those backup generators, depleted batteries, etc. how much more likely it is the "grids" will have some little bugs floating around somewhere that cause the "infallible" system to fail.
    The concepts are great as concepts go. But I think anyone should approach such decisions with a grain of salt (or common sense) that any system has the potential for problems, the more complex the system, the greater the potential. I wish these folks well, but after all; the multi-billion dollar Microsoft still releases patches & fixes all the time and multi-million dollar datacenters can be brought down by a 50 cent part so some problems are to be expected when undertaking such a major task.
    And all the testing they may have done prior to going live can't replicate what a thousand actual users will do (since all hosts know customers may do anything and probably will).
    So for now I still contemplate that that old idea of having multiple "simple" "tried & true" conventional server set-ups (not $59 Celerons but not super-complex "systems") may still offer less potential for masses of upset customers than one "super system" where one unforeseen point of failure can possibly cause everybody problems.

    Just my 2 cents & thoughts on the "whole idea". Although even as a host myself the concepts are interesting enough if I ever get time I might try an account on one of these type systems just to see what they do and get an idea what everybody here is talking about.
    New Idea Hosting NO Overselling-Business-Grade, Shared Only! New-In House Design Team.
    High Speed & Uptime; , DIY Pro-Site Builder-Daily Backups-Custom Plans, All Dual Xeon Quad Intel servers w/ ECC DDR3 RAM SCSI RAID minimums.
    We Concentrate on Shared Hosting ...doing one thing and doing it VERY well
      0 Not allowed!

  25. #25
    If you want to know if it is stable yet then i suggest doing this:

    http://blogsearch.google.com/blogsea...TF-8&scoring=d

    You'll see people's blogs where they complain about their hosting :-/
    Last edited by RossH; 11-16-2006 at 01:21 AM.
      0 Not allowed!

  26. #26
    DDT - I must say - this is one of the better posts I have read on WHT in quite awhile !!!!!

    I am not certain I can improve on this - nor would I want to try - however, I would like to add to your comments if you dont mind

    Things like H-sphere and newer load balancing & clustering set-ups (and more advanced versions like the grids) may be changing that perspective some, but I can't help but still wonder about that basic idea.
    You are exactly right with this statement. I am a HUGE proponant of clustering and specifically utilizing hsphere - however, this alone does not guarantee uptime. The powerful benefit of hsphere is that you are spreading services across multiple servers - this is not a High Availablility Solution per se - it is just distributing tasks across multiple servers - however, if done properly, it certainly yields higher uptime and performance results over single server silo'ed environments....

    The author's "food for thought" was "is it better to have a monster system that if it fails or slows to a crawl results in tons of mad clients, or is it better to have a larger quantity of more "standard" servers so if 1 fails or slows down you only have the support issues from the people on that one box"?
    And here in lies the problem - and the problem is that people - at least on this forum do not have a clear definition or understanding of what high availability actually means

    High Availablility is a strategy with a target objective - whether that is 99.9%, 99.99%, 99.999%, etc - the strategy involves identifying single points of failure and creating a "gameplan" for addressing incidents which will enable you to maintain your stated objective

    No one in their right mind promises 100% uptime - it just does not, and cannot exist...

    IBM offers, at their highest level - 6 9's (ie 99.9999%) - and the cost for such a solution is in excess of 1M/year and it does not come anywhere near to promising 1TB of transfer/month for $20 or $100 or whatever - this just isnt possible -

    you may still have single points of failure
    OK - so, lets define a single point of failure....

    If you have a web server in a cluster with redundant hard drives, raid, dual power supplies, dual fans, dual nics, etc - the single point of failure on that server is the motherboard....

    It could be that the success of their marketing department really exceeded all their expectations.
    Dan, as usual - you have gotten to the core of the issue with a simple sentence...

    The ultimate problem is overselling - and I know everyone on this forum thinks its normal and everyone does it - but, what everyone is missing is that overselling creates additional single points of failure...

    Overselling cannot be planned for - and hence, it cannot fit into a HA plan... At any point - 1 or more domains can experience situations which are unpredicatable and can affect your uptime...

    Lets assume that a server can handle 500 units (what these units actually mean is irrelavent)

    so, you have a "high availability" cluster of 4 servers where web servers are load balanced so that all incoming requests are shared amongst the 4 servers - the theory is that if 1 server goes down, the others will take over and hence - zero downtime...

    Well, what happens if the total load you are asking of these servers is 2000 units or higher - if 1 server fails, then the load balanced array will fail - therefore, you have now created 4 single points of failure...

    So, a single undersold server, may in fact be more reliable and may be more viable for a high availability strategy then a load balanced node of multiple servers...

    I have often maintained - and can back up by actual results - that a services cluster is more reliable then an oversold, load balanced cluster - as there will be less single points of failure.. and an undersold load balanced cluster will be even yet more reliable...

    the problem is, everyone is chasing the dream of 100% uptime, 1000 GB transfer accounts for under $100 - people - this isnt going to happen - at least not in the forseeable future...

    I also must comment on something at this point. I am a HUGE fan of what MT is doing. I think they are bang on with their strategy and bang on with their concept of CPU utilization - I have heard their podcast and am extremely impressed with their direction, professionalism and understanding of the market place - I just wish they had gone to market with a different strategy - there is NO reason to get themselves into a marketing gimmick price war - they had a strong enough value statement - and I believe they should have had more confidence in their offering and maintained a price point that was attainable for their objectives...

    If they add enough nodes to their system, they will stabilize - and I hope they do so - however, as long as they are offering 1000 GB of transfer for $20, they will constantly be in emergency mode - and really, they dont need to be...
    Last edited by cartika-andrew; 11-16-2006 at 02:38 AM.
      0 Not allowed!

  27. #27
    Join Date
    Oct 2006
    Posts
    51

    Some Good Rational Thinking

    Quote Originally Posted by DDT
    As I have said in other posts the "Grid" concept is interesting but doesn't seem to have the bugs out.
    Then you have load balanced clusters etc.
    However I keep being reminded of an interesting article I read a long time ago (before all the "grid" stuff) about as a host grows whether to go with huge servers & bigger RAID arrays or clustering or whatever. The jest of the article (and I understand that conceptually the "grids" are supposed to eliminate this but obviously it's not all smooth sailing yet) was that with big servers or clusters etc. you may still have single points of failure (even if it's just a switch) The author's "food for thought" was "is it better to have a monster system that if it fails or slows to a crawl results in tons of mad clients, or is it better to have a larger quantity of more "standard" servers so if 1 fails or slows down you only have the support issues from the people on that one box"? Things like H-sphere and newer load balancing & clustering set-ups (and more advanced versions like the grids) may be changing that perspective some, but I can't help but still wonder about that basic idea.

    I'm sure these folks have teams of admins pulling their hair out trying to get everything running smoothly. But these are still just hosts. Recently I read another article (I guess I like to read about other people's problems ) where an entire floor of servers in a new state-of-the-art super-redundant datacenter was brought down by a power failure that "rolled" through all the redundant systems. Their on-site highly skilled techs found the problem and used "jumpers" to get things on-line then a high-priced failure forensics team came in to find out what the heck really happened. It all came down to a little electrical connector in one circuit box, not even a $1 part. This led to analysis of the entire center where they found 20 some of these little connectors in various locations had the same corrosion problem and could potentially fail. This led to thousands of dollars in replacing all these little connectors in all the boxes in the facility. Evidently a "bad batch" of connectors the electrical contractor was supplied. My point being if this new-state-of-the-art center could be brought down by one little connector resulting in the malfunction of all those backup generators, depleted batteries, etc. how much more likely it is the "grids" will have some little bugs floating around somewhere that cause the "infallible" system to fail.
    The concepts are great as concepts go. But I think anyone should approach such decisions with a grain of salt (or common sense) that any system has the potential for problems, the more complex the system, the greater the potential. I wish these folks well, but after all; the multi-billion dollar Microsoft still releases patches & fixes all the time and multi-million dollar datacenters can be brought down by a 50 cent part so some problems are to be expected when undertaking such a major task.
    And all the testing they may have done prior to going live can't replicate what a thousand actual users will do (since all hosts know customers may do anything and probably will).
    So for now I still contemplate that that old idea of having multiple "simple" "tried & true" conventional server set-ups (not $59 Celerons but not super-complex "systems") may still offer less potential for masses of upset customers than one "super system" where one unforeseen point of failure can possibly cause everybody problems.

    Just my 2 cents & thoughts on the "whole idea". Although even as a host myself the concepts are interesting enough if I ever get time I might try an account on one of these type systems just to see what they do and get an idea what everybody here is talking about.
    Apart form maybe not Knowing that DDT is a banned chemical (Canadian humor - sorry), you make some valid points that I wholeheartedly agree with. Your reasoning and explanation are well thought-out.

    Thanks you very much Mr. DDT for taking the time to analyse and comment.

    Les (In Canada - Where else?!?!)
      0 Not allowed!

  28. #28
    Join Date
    Oct 2006
    Posts
    51

    Tech Talk Got Me Also!

    Quote Originally Posted by marak
    You're definetly not bashing them. They unleashed their PR monster, making arrogant comments like "The grid won't go offline until California goes into the ocean", and all these other things. After that kind of PR, and all their 101% uptime and gimmicks, they should provide what they say. I think we should continue to comment until they get their act together.
    Ha - You listened to that TechTalk Padcast too! Man, Dumb**s that I am - I fell for it - hook line and sinker.

      0 Not allowed!

  29. #29
    Join Date
    Jul 2006
    Posts
    74

    Pr

    Ha - You listened to that TechTalk Padcast too! Man, Dumb**s that I am - I fell for it - hook line and sinker.
    I fell for it too... The only difference was I had to OK it with my boss, and in the meantime, a colleague ran into one of the threads. Personally, I am looking at the grid layer.... They provide email support only, but really, how much real support can actually be done over the phone.
      0 Not allowed!

  30. #30
    Join Date
    Oct 2006
    Posts
    51

    Good Souls

    Quote Originally Posted by marak
    I fell for it too... The only difference was I had to OK it with my boss, and in the meantime, a colleague ran into one of the threads. Personally, I am looking at the grid layer.... They provide email support only, but really, how much real support can actually be done over the phone.
    Well here's something to cheer you up:

    http://marak.youaremighty.com/

    Have good 1!

    BTW, tell your boss "the GRID" is getting better --- bit by bit.

    Les
      0 Not allowed!

  31. #31
    Join Date
    Jul 2006
    Posts
    74

    Better and Better

    les,
    Thanks for the link.
    BTW, tell your boss "the GRID" is getting better --- bit by bit.
    It's good to know that they are improving. Keep us updated; this will be a great solution if they can get it to work.
      0 Not allowed!

  32. #32
    Join Date
    Aug 2005
    Location
    Canada
    Posts
    838
    Tera3 (the company providing the Applogic OS to GridLayer) confirmed they have nothing to do with MT.
    http://www.webhostingtalk.com/showthread.php?t=558772

    They seem to have pretty ambitious vision.
    Very interesting.


    But I hope MT to get things together and compete with them, too.

    It's better to have a bit of competition.
    It can stimulate both, and make both of them better.
      0 Not allowed!

  33. #33
    Join Date
    Dec 2004
    Location
    New York, NY
    Posts
    10,574
    Quote Originally Posted by extras
    Tera3 (the company providing the Applogic OS to GridLayer) confirmed they have nothing to do with MT.
    http://www.webhostingtalk.com/showthread.php?t=558772

    They seem to have pretty ambitious vision.
    Very interesting.


    But I hope MT to get things together and compete with them, too.

    It's better to have a bit of competition.
    It can stimulate both, and make both of them better.
    For some reason I really don't see TheGridLayer as competition to MediaTemple...
    MediaLayer, LLC - www.medialayer.com Learn how we can make your website load faster, translating to better conversion rates for your business!
    The pioneers of optimized web hosting, featuring LiteSpeed Web Server & SSD Storage - Celebrating 10 Years in Business
      0 Not allowed!

  34. #34
    Join Date
    Aug 2005
    Location
    Canada
    Posts
    838
    MT has intention of going into resller market with their Grid solution.
    Although GridLayer won't probably provide account for end users, they are supplying Grid nodes for people who would be doing that.

    So, eventhough not necessarily they are in the same market (in the current marketing model), they could be competing (especially) in the "Grid" market, IMO.
    (We don't know what kind of "Grid" MT is using yet, though.)
      0 Not allowed!

  35. #35
    Join Date
    Nov 2006
    Location
    Southern California
    Posts
    173

    *

    Quote Originally Posted by extras
    MT has intention of going into resller market with their Grid solution.
    Although GridLayer won't probably provide account for end users, they are supplying Grid nodes for people who would be doing that.

    So, eventhough not necessarily they are in the same market (in the current marketing model), they could be competing (especially) in the "Grid" market, IMO.
    (We don't know what kind of "Grid" MT is using yet, though.)
    Well, I can't speak for the GridLayer folks - I don't know what they will decide to offer as services/products from their grids. However, if they use the full capabilities of AppLogic, they should be able to offer at least the following:

    Basic infrastructure:
    ------------------------
    - grid servers (available today)
    - grid clusters (firewall + load balancer + 2-8 web servers + database server + NAS, coming soon)

    Web hosting infrastructure:
    ----------------------------------
    - Cpanel clusters (in Beta now)
    - Ensim clusters (coming soon)
    - H-sphere clusters (coming soon)

    These will be completely integrated and allow you to manage up to 16-32 servers in a fully transparent & load-balanced fashion (like having a single huge server with up to 64 CPU, 128 MB RAM and 10TB of disk space)

    Application infrastructure:
    --------------------------------
    - LAMP clusters (available today)
    - J2EE clusters (coming soon)
    - RoR clusters (coming soon)

    These are also completely pre-integrated - just add HTML/XML, code & database. Useful for Web apps / SaaS apps like SugarCRM, WordPress, etc.

    Grid infrastructure:
    -----------------------
    - shared grid accounts (you get to build, customize & control your infrastructure like on a dedicated grid, but run it on the shared grid)
    - dedicated grids (available today)

    All of these can be served off the same grid, and will inherit the same key advantages:

    Instant provisioning
    -------------------------
    With a proper portal, you can pick a cluster, pay by credit card and have it running on the grid in 5-10 minutes)

    Universal high availability
    -------------------------------
    When something crashes on an AppLogic grid, whether a physical server or a virtual appliance, the data remains available at all times and the affected appliances are restarted instantly somewhere else on the grid without human intervention. This means that ANY server or cluster running on the grid is highly available. Today, there is an up to 4min delay when a physical server dies; with a bit of hardware assist, this can be reduced to under 30sec.

    Instant scalability
    ----------------------
    ANY AppLogic infrastructure scales in at least two ways: (a) you can change CPU, memory, bandwidth and disk space, and (b) you can add/remove servers from a cluster. Both are done on your exising server/cluster, without requiring any data/code migration. Each can be done in under a minute if GridLayer implements a web portal that lets people control their grid accounts directly rather than through the ticketing system.

    Private software/configuration
    --------------------------------------
    One thing to keep in mind is that in AppLogic, only the hardware is shared - while you are running on a grid, you always get your own private firewalls, load balancers, web servers, app servers, mail servers, database servers, etc. This means that there is simply no way for someone else's load to affect your operation. Also, there is no way for the host to oversell - AppLogic enforces resource quotas rigorously.

    Utility pricing
    ----------------
    AppLogic has a metering system which can tell the hosting provider who consumed how much resources and when. This is important because it allows the hosting provider to let customers adjust their resources directly - you can boost up CPU/memory/storage/bandwidth when you need them, and cut down when you don't.

    The economics of using AppLogic vs. homegrown grid implementations is also very different. AppLogic does not require ANY expensive hardware (no firewalls, no load balancers, no SAN - just a bunch of servers connected by Gigabit Ethernet). And, you can use older servers that have already paid for themselves...

    So, I am not sure how MT or Mosso will be able to compete. They look just a shared web/PHP hosting environment stretched on a grid...
      0 Not allowed!

  36. #36
    and it's down once again.....
      0 Not allowed!

  37. #37
    Join Date
    Oct 2006
    Posts
    51

    MT Grid Server is down

    Quote Originally Posted by RossH
    and it's down once again.....
    Yup! No sites - no Account Management - NO FTP - pretty soon no clients!

      0 Not allowed!

  38. #38
    Join Date
    Oct 2006
    Posts
    51

    Why Media Temple's Grid Server Service Went Down Today

    Quote Originally Posted by joomlales
    Yup! No sites - no Account Management - NO FTP - pretty soon no clients!

    Just in case people want an explanation

    LOL - Les
    Attached Thumbnails Attached Thumbnails MT Issues.PNG  
      0 Not allowed!

  39. #39
    Join Date
    Feb 2005
    Location
    Curitiba, Brazil
    Posts
    123
    They should have a beta test with real clients warned and given a discount or a bounty. Like Google does for its products.

    I still can't believe the difference between MT's 100GB/1TB @ $20/m offer compared to The Grid Layer's 10GB/1TB @ $50/m offer. Any explanation why TGL offers so little space?
      0 Not allowed!

  40. #40
    Join Date
    Jun 2004
    Location
    Bay Area
    Posts
    1,320
    Because TGL's offer is not extremely overbooked, has guaranteed resources and such?

    Oh, and its actually stable
    Last edited by Xandrios; 11-21-2006 at 12:08 PM.
      0 Not allowed!

Page 1 of 2 12 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •