For some context -- Slicehost has had sort of a user insurrection on their forums, complaining that despite their massive influx of capital from Rackspace, Slicehost is totally uncompetitive with Linode and the hundreds of other nearly identical, commoditized hosts.
The creator has been apologizing for many months acknowledging that Slicehost has been falling signficantly behind and has stagnated. However, he has been saying essentially "Don't give up yet because February this will all change."
This is the February announcement. Basically you get slightly more bandwidth.
Funny, I always use Slicehost as the example of a non-commoditized hosting arrangement: back when you could get disk space and terabytes of (gratuitously oversold and capriciously available) bandwidth for $2 a month, they charged ten times that for 256 MB of RAM that would actually be there when you wanted to use it, plus really outstanding support and reliability. Some years later they've got competition in their space, but I still see outstanding support and reliability.
I wouldn't mind an extra gig of RAM and a pony but I'm not fixing to move anytime soon.
Have fun! I am going to take my virtual disk image to another service, and then switch back to Slicehost when they decide to start competing again. :)
I didn't mean "commoditized" in a bad way, I mean in the sense that Slicehost is doing something that a hundred of other companies are doing as well (running Xen or equivalent on a datacenter and selling virtual computers). It is easy to benchmark and scientifically see which service outperforms another on any given metric.
There is absolutely no reason to stick with Slicehost when you can switch to another datacenter in under an hour, and then switch back the week after if you'd like (prorated, of course).
You really should try out linode for a day and even if it comes out equal to slicehost in the end you really have no reason not to switch as it will get you more for the same price. (Unless of course your server setup is so immensely complicated that you can't afford the time it would take to setup a new one.)
Its a question of priorities: I pay $150 a month for hosting. Next to my sales, that's mouse droppings. Getting an extra 512 MB of RAM or 1 GB of RAM at the same price is pretty meaningless to me: the app runs. Running on more RAM will not help me sell more software.
Hypothetically saving $50 a month on hosting is also less than motivating to me -- "Try a new host out for a day" means I lose a day that could be sent implementing things that will increase the amount of software I sell in a scalable fashion. For example, A/B testing (or SEO, or writing better email copy, or...). Note that future improvements to cost reduction are not multiplicatively effective but future A/B tests (etc) are. (Saving $50 on hosting costs is a one-off improvement -- it doesn't make my next improvement more effective. Increasing conversions by 1.3% anywhere in my funnel right now would make me about $50 a month. A 1.3% lift does make my next improvement more effective because funnel conversions are essentially multiplicative: 10% improvement at gate A and 10% improvement at gate B means 21% improvement total, not 20%. Most of you probably already know this but if it is news to you bookmark this factoid and come back to it later because it is really freaking important.)
Getting an extra 512 MB of RAM or 1 GB of RAM at the same price is pretty meaningless to me
I wouldn't mind an extra gig of RAM and a pony
The second statement highlighted is why I said you should try linode for a day. To clarify what I meant is you should try linode out (not as a production server) but to see if you can use it the same for that extra gig of ram for the same cost or less with minimal effort.
I really wasn't trying to say it would save you money in the end or that it would help sell software as you seem to think.
Edit: and btw your Buy Now button is still really hard to see as I commented before
I really wasn't trying to say it would save you money in the end or that it would help sell software as you seem to think.
I think you might be missing the point here. If it does not in fact save significant money or help sell software--he's not interested.
I have a (currently underused) slicehost account. In my opinion these guys have historically been known for awesome customer service. They need to continue to do that, and not try to compete on price. I am not optimistic about their ability to deliver amazing service after the Rackspace acquisition, though.
I don't think it's the owner that was saying "wait until February," but rather just one of the other employees - one that wrote many of the original tutorials and was hired on because of that.
It was PickledOnion (Paul) who made the claim about announcements coming in February. He was promoted to a management position (he refers to himself as their "GM" in this thread: http://forum.slicehost.com/comments.php?DiscussionID=4348...) and appears to be the, or one of the key decision makers at Slicehost now. In that thread, he says that he accepts responsibility for Slicehost languishing during the course of the last year.
I think it's time to pack my bags and go support Linode or Prgmr for a bit. Slicehost is clearly in a comfortable place, and I'd rather help spur on the competition.
We transferred from Slicehost to Linode a few months ago. I believe that the Rackspace acquisition turned Slicehost into a zombie company that is not being run with the same common sense attention to care I was seeing when I originally became a customer.
My personal hope, now that I'm almost finished transferring my main site to linode, is that they reject buy-outs and play to win. HostGator has done this, and they seem to be doing extremely well. If they had offered VPS when I was with them, I would still be a HostGator customer. I would love to be able to stick with a host for more than ~18 months. With linode's flexibility, this seems like a strong possibility.
There are so many reasons I am glad I started and stayed with linode for a VPS. All I hear about slicehost anymore is that they fall under linode in benchmarks, they require 64bit OSs (don't know if this has changed), and they fail to meet other VPS offers.
Personal experience: Linode move from a linode 360 to a linode 540 took under 15 minutes from time submitting support ticket to resizing and rebooting, pro rating pricing is very awesome
Sad to see them go. I guess it was inevitable when Rackspace bought them.
They should have matched Linode's offer. Perhaps some additional features to entice Linode customers over to Slicehost. Instead we get more packages and a slightly more dreary website. That's not competition.
I thought this was more of an issue with the arc language setup than with the host. I remember reading something about having to "reboot" the system every few hours.
Having played a small part in the general migration to Linode ( http://journal.dedasys.com/2008/11/24/slicehost-vs-linode ) , I think it's unfortunate that Slicehost doesn't up their game a bit more. I actually started with them after moving away from Layered Tech, and Slicehost are good, competent people with a good product. They wouldn't be that far behind Linode if they'd ditch the "64 bit only" thing. They don't seem to want to change though. I wouldn't have even written that article if they had said they were working on 32 bit support.
Also, I hope that the influx of people to Linode doesn't degrade their ability to do the good job that I've experienced so far.
I'm a long time Slicehost customer, and really like them. I like the new site layout. If the layout is whats stopping you from being a customer, then I'm not sure why anyone would use Prgmr. Basically, its usable, so I see no reason to complain.
I wish the big announcement was that they were doubling their disk size or some sort. I've had websites die because a valid MySQL connection couldn't be made because the slice ran out of disk space.
Edit: Oh wow, while I was typing this up, they changed back to their old design.
Edit #2: Double wow, Prgrm's prices are amazing. I think its time to check them out.
The differences are not minor, though. 20$ gets you 256M on Slicehost, 360 on Linode. You also have to factor in that you're forced to use 64 bits on the Slicehost system, so you get less bang for your buck.
You're only forced to use 64 bit userspace on slicehost if debootstrapping a 32 bit chroot and replacing the root filesystem with it live is too scary. :)
The real nice thing about linode is they allow running your own custom linux kernel.
Even were it a straight comparison, Linode is still cheaper, so why would I pay for the privilege of fooling around every time I want to set up a new instance?
If someone writes a tutorial, more people would do it. I wanted to when I was on Slicehost, but ended up just switching to Prgmr instead because it was easier.
I started using slicehost just about when they were bought by Rackspace. While I can only praise their services, the prices have become from competitive to obviously premium. Also they have no servers in Europe, which really hurts me.
I've started moving to a local company, www.intovps.com. They don't offer the same stability (I had a sporadically bad connection for about 4 days before it was fixed), but the prices are exactly 1/4 and performance is good.
I'd probably move completely if I didn't have a lot of DNS records with slicehost. Anybody knows how I could migrate them from slicehost to linode or somewhere else?
Why is everyone complaining about a site redesign? As far as I can tell it looks exactly the same? Was the CSS file not included or something so everyone freaked out? I'd love to hear some context on all of this...
No offence, MediaTemple sucks. They can be more cost effective but their uptime and performance is absolutely horrible. I almost lost a big client because of their constant outage and performance issues. I learned my lesson and moved out.
The only thing I can't comment on is their Dedicated server services because I haven't used it. But honestly, there are better DS out there with almost half of MT's price with same configuration.
Are you talking about the GS? I had a lot of trouble with it when I tried it, but decided to try the DV before giving up on them. It's been a world of difference - it's been rock solid, fast, and seemingly invariate in performance.
The prices listed for the DVs might seem high, but they've been good about bumping up RAM allocation for free. The amount listed isn't the amount you actually get, and with the free extra, it's more cost effective than Slicehost, as well as Linode (at least at the 3 GB pricepoint).
I have used both DV and GS. DV for clients and GS for personal use. Both horrible.
For dedicated servers I used hostgator. It was good, but they can't match the price of Slicehost/Linode VPS price (last time I checked).
To be honest, unless you are running a HUGE project (a la digg/reddit/twitter), slicehost/Linode type VPS are more than enough. You can start small and slowly scale as you get bigger. Their largest VPS will meet the needs of most medium sized web business out there.
I second this. I am moving off of their $150/month Rage DV to a couple of linodes. I started with a $50/month Base DV, then quickly upgraded as I started hitting memory limits, and finally went to Rage when I was having I/O thrashing issues (even during my site's "off hours"...). That didn't help at all. Since the next step up at MT is a $750/month Nitro, I was left with two more palatable choices: add another DV, or jump hosts.
I'll make an update when I finish the transfer to linode and figure out exactly what sized linodes I end up needing. I'll also let people know if this was simply my fault (bad coding).
Hm, maybe I just got lucky with the server my DV is on (or maybe you were unlucky). Regardless, I think they really need to get their GS act together, though... it was by far the worst hosting experience I've ever had.
But my original point was that it's sad that my MediaTemple DV is a better value than Slicehost now... MT is not exactly a value host.
Given so many fine gradiations of slice-size, it'd be interesting to see an analysis, for distributable workloads, when it's better to use more smaller slices vs. fewer larger slices of similar cost. (On what sort of jobs will 12 256slices for $240 beat one 4096 slice for $250, and so forth?)
I spoke to a Rackspace Cloud Servers Engineer a few months back, and he said that they're the same servers (same specs), in the same data centre. Not sure if he's 100% accurate, but regardless, as the customer, the only differences I see are different control panel design, inclusion of bandwidth in price vs "pay as you go", and a monthly vs hourly billing.
The creator has been apologizing for many months acknowledging that Slicehost has been falling signficantly behind and has stagnated. However, he has been saying essentially "Don't give up yet because February this will all change."
This is the February announcement. Basically you get slightly more bandwidth.