New Hardware and status on lag and invoker

[Whirlin]

Building your own server, much like someone building their own computer, leaves you to deal with each part manufacturer separately if something breaks. In a home environment this is usually fine because you are accepting your own personal level of fault tolerance and downtime. In a business, you are impacting your bottom line and your customers. The parts described could easily be replicated in a very cost-effective build from HP or Dell with a matching 3 year + 2 hour response time onsite warranty, meaning if something breaks they will usually be able to get a replacement part onsite the same day and get things going again. I also don't see, but this doesn't mean it doesn't exist on the box, any out of band support access such as ILO or IDRAC that allows someone to remotely manage a server even if turned off.

Everything you stated is absolutely true.

However, I think you're downplaying the costs of the hardware and support, overstating the benefits. The conversion of a one-time build charge to not only have a more expensive initial build through a particular party, but then adding in a continued maintenance cost. The next-best-alternative to the hardware contracts by having a GM within range to provide onsite support, which they already have on their payroll. With the comparative costs, they could afford having a few additional components on hand and configured to swap in case of a particular drive outage. Once again, you're right, this is cheaper on two fronts... and I realize the margins are thinner now than they used to be 5 years ago, but boutique computers, especially in the enterprise space, can run at 2x on the initial build, with 5-10% additional per month depending on service provider on an ongoing basis. The difference in costs could be used to reserve a few additional components for on-site support. Additionally... support isn't always supporting and helpful.... I'll just leave it at that.

Secondly, your statements also imply there's no redundancy in either the physical or logical layers, and any individual component crash results in an outage or loss. I'd be surprised if this is the case with seemingly identical architectures.

Ideally though, your assertions are 100% spot on... however, I think in this particular use case, their position is defendable.

[Orthin]

Everything you stated is absolutely true.

However, I think you're downplaying the costs of the hardware and support, overstating the benefits. The conversion of a one-time build charge to not only have a more expensive initial build through a particular party, but then adding in a continued maintenance cost. The next-best-alternative to the hardware contracts by having a GM within range to provide onsite support, which they already have on their payroll. With the comparative costs, they could afford having a few additional components on hand and configured to swap in case of a particular drive outage. Once again, you're right, this is cheaper on two fronts... and I realize the margins are thinner now than they used to be 5 years ago, but boutique computers, especially in the enterprise space, can run at 2x on the initial build, with 5-10% additional per month depending on service provider on an ongoing basis. The difference in costs could be used to reserve a few additional components for on-site support. Additionally... support isn't always supporting and helpful.... I'll just leave it at that.

Secondly, your statements also imply there's no redundancy in either the physical or logical layers, and any individual component crash results in an outage or loss. I'd be surprised if this is the case with seemingly identical architectures.

Ideally though, your assertions are 100% spot on... however, I think in this particular use case, their position is defendable.

Well thought out and meaningful responses have no place round these parts feller’ this here’s the PC!

[Berost]

Well thought out and meaningful responses have no place round these parts feller’ this here’s the PC!

[Shaps]

Berost I love your profile pic. Now that is a Dwarf. /salute

[AnOrdim]

Everything you stated is absolutely true.

However, I think you're downplaying the costs of the hardware and support, overstating the benefits. The conversion of a one-time build charge to not only have a more expensive initial build through a particular party, but then adding in a continued maintenance cost. The next-best-alternative to the hardware contracts by having a GM within range to provide onsite support, which they already have on their payroll. With the comparative costs, they could afford having a few additional components on hand and configured to swap in case of a particular drive outage. Once again, you're right, this is cheaper on two fronts... and I realize the margins are thinner now than they used to be 5 years ago, but boutique computers, especially in the enterprise space, can run at 2x on the initial build, with 5-10% additional per month depending on service provider on an ongoing basis. The difference in costs could be used to reserve a few additional components for on-site support. Additionally... support isn't always supporting and helpful.... I'll just leave it at that.

Secondly, your statements also imply there's no redundancy in either the physical or logical layers, and any individual component crash results in an outage or loss. I'd be surprised if this is the case with seemingly identical architectures.

Ideally though, your assertions are 100% spot on... however, I think in this particular use case, their position is defendable.

I went through the portals I have to spec stuff out and was having a hard time even finding a true rack mount server with that particular processor. It is mostly available in workstation configurations as the name implies. So I'll give them the benefit of the doubt and say they did their homework and determined that the best-case scenario for performance was the highest clock speed processor they could get within a certain price range. Compared to the Bronze/silver/gold/platinum level Xeon "scalable" server processors, the W-2245 has a much higher core and boost speed at a fraction of the cost. There certainly are a lot of unknowns, but let's say they were to employ a clustered environment. I think that just reinforces my point, particularly with the consumer-grade SSDs. The controllers are not designed for proper RAID features and while yes they will probably work, that doesn't mean you should be doing it that way. That's not even getting into if they actually bought a proper RAID controller card or are just going to use the software RAID controller that comes with the board. I'm not seeing a capable RAID controller on that board anywhere. We also have little to no information, that I have access to anyway, on what OS they will be running etc to know if clustering is possible to provide any redundancy.

To be clear, I think the hardware is fine on paper and would be a perfectly serviceable garage server or testing environment for a business. But I would never put something like that into production let alone an internet-facing service that generates revenue. Even the budget parts total can be argued to not be "cheap" but when you are charging a subscription with multiple tiers on top of frequent direct-purchase events I would have expected a little more invested into the foundation of the house. This is all coming from the same company that has consistently been unable to replace their SSL cert in a timely fashion before it expires.


You stupid piece of shit.

[Taernath]

You stupid piece of shit.

[AnOrdim]

The peanut gallery was getting restless so I tossed them a bone.

[Methais]

Well thought out and meaningful responses have no place round these parts feller’ this here’s the PC!

QUICK, CALL HIM A DICK OR SOMETHING BEFORE IT'S TOO LATE!!!!

EDIT:

You stupid piece of shit.

Nevermind.

[Stanley Burrell]

Dude, I was put into 20-second RT because when SIMUCOIN DELIVER'ing + receiving Store items was (according to the GMs) creating an inordinate amount of screamers. I don't want to upset the powers that be, so I've been doing CONSIDER {self} twice, between each delivery, lol. I don't want to give them a headache, but maybe this was one of the things, discussed, that will hopefully be fine-tuned. Alright.

As a former member of staff, set off all the fucking screamers you can. They need to fix that shit.

Ahhh, not so sure I want to set off all the screamers (I have a few that SORT with my buggy Skellie absolutely makes miserable for the higher-ups.)

And of course, the time I was trying to swim into Hisskra's and while under water, swimming was causing screamers because I had a Paladin-bonded weapon on me and I was like, "lol. Time to drown."

I would be curious to *know* which ones are still there, but I don't want to be thrown into the abyss by the GameMasters. And then be eaten by a Grue. That's how that happens I think.



Also, I fucked up the quote tag placement in this post and am too lazy to fix it. Bye.

[mgoddess]

I would be curious to *know* which ones are still there

I've heard the morphing runestaves (that can morph into wristlets, belts, etc.) are broken enough to send screamers out... I'm tempted to try, but don't want the GM-slapdown. heh