New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Their support and infrastructure is not close to being comparable to DO. You will never see the five dollar Linode plan. They won't race to the bottom and they deliver a premium product. You startup in DO on the cheap then move over to linode when you're serious
Not to be disrespectful, but I think you are the one who doesn't get it.
First, 4x as many support tickets, 4x as many billing inquiries, etc.
Second, it's impossible hard to make money on a $5/month VPS given the amount of support Linode promises.
And finally, your theory of exposure and tradeup is not universally true. In some markets having a spread of prices works well, but not in all. To borrow your theory - why doesn't Apple offer a $299 MacBook? Why doesn't BMW offer a $9,999 car? Some companies are Fiat and some are Mercedes. Just basic market segmentation - you can chase the broad market or the high end.
The people for whom the difference between $5 and $20 a month for a VPS is significant are not clients Linode is interested in having. There are lots and lots and lots of people who are happy to pay $20-40/month for a VPS as long as they get some support and it works well. That's Linode's market.
As much as I like DO, if I needed the 2GB plan I will be at Linodes door first.
Uh-oh, this sounds so familiar. Many many technology companies said "it's ok, we'll just focus on the premium segment" -- and perished as soon as their "non-premium" competitors started delivering product that was mostly good enough for most of their customer base. It's called commoditization and the most notable example is the rise of the "personal computer" x86 architecture in the server space and downfall of ye olde "Big iron", the super-reliable but uber-expensive high-end hardware based on DEC Alpha, SPARC, SGI MIPS etc. Also plenty of similar examples in the software market as well.
You're missing the point. They are not targeting every Tom, Dick, Harry and skiddy on the face of the earth. Target markets are a pretty basic business premise.
It makes sense when you think about it from the standpoint of a business, rather than a cheapskate.
That sounds really promising . . . without any actual analysis or statistics, anyway . . .
My prediction....
Linode will retain the $20 minimum entry point. But in the future you'll be able to divide your resources, e.g., 2 x 1GB servers. Like a few providers are doing now...
Wonder if DigitalOcean lowers prices or bumps up specs now. The main reason to go with DO over Linode was price. I would never use the $20 and up DO plans when I can get equal or better on Linode.
That's what's going to be interesting to watch, and not only DO.
As someone else pointed out, I would have expected the SSD jump months ago.
And most of the commodity technology companies also died or pivoted. HP, Compaq, Gateway, IBM (Lenovo). Compaq absorbed DEC, which ended up losing and having to merge with HP. HP's PSG group has pretty low margins and as a corp, relies heavily on their enterprise and services segments. It has a self-interest to promote the PSG, but trying to do both has taken a toll. IBM divested Lenovo to a low-cost manufacturer/supplier and themselves moved to 'premium'. Gateway almost died several times as it absorbed the mind-share company eMachines. Only to be bailed out again by a supplier/manufacturer, Acer.
I dunno, your example only teaches me one thing: If you're going to dabble in LEB, then you better control the manufacturing line and be housed in China.
Linode doesn't even need it. My non-SSD linode gets 750MB/s + running everyone's favorite Linux command.
Upgrading now just to see if there is a difference.
EDIT. Just upgraded....
dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 1.42948 s, 751 MB/s
No difference. Had the same disk speeds before. Now I just have more RAM and less CPU.
Wondering how they would handle 10gbps attacks on a linode box, now that it theoretically can receive 40gbps.
Well, why it still worse than DO by spec then?
I can't see my old linode.
I bet there is, especially in high IOPS usage situations. What you did probably took place in cache, hence the no difference thing.
Because spec always determines performance? I bet Linode are not using the cheapest desktop grade SSD's that you will find in most providers boxes, the list goes on to be honest.
Welcome to 2014 Linode, you are a little late.
I tried that 10G+ - Inbound no chance, i max out at not even 1500Mbit no matter what sources (3x Leaseweb + Tele2) i use
As much as i like Prometeus/iwstack for smaller plans, when i need the 2GB plan i just stay with Prometeus. But i get your point.
On the 96GB plan? I wouldn't be surprised if there were hiccups on the new network but all the plans have different port speeds
All are 40 Gbit inbound, which is what he tested...
I also find it hard to believe each of their hardware nodes actually has 4x10GbE NICs in it.
(Or a 40GbE NIC? Which could cost more than the rest of the node itself?...)
You're right on that, my bad, but it does say:
Which wouldn't leave it out of the realm of possibility that each plan has a different inbound speed, it does say up to just like on the outbound.
No, new Supermicro boards have 1-2 40Gbit QSFP onboard replacing the old 2x 10GBE SFP ports.
Switching/Routing is expensive unless you use older Force10 refurb gear but i doubt this few hundred k are an issue for Linode.
Actually i couldn't find any 40Gbps ethernet solutions from SuperMicro, just Infiniband.
Are Linode running their new network over Infiniband? Interesting.
https://blog.linode.com/2013/03/07/linode-nextgen-the-network/
@bdtech that blog post there talks about "bonded gigabit links" between the host nodes and the fabric extenders. Then there are 10G links from the Nexus 2000 fabric extenders to the Nexus 5000 switches, reaching total of 40G to a fabric extender if their load balancing works.
This is very different from 40G to each host node.
The reason of this update:
I was linode client in 2012 and 2013. (around 2 years).
Linode have in compare between other ISP, only good wiki, and fast support answers.
In 90% cases these support using a common phrases.
Nothing more about them i can't to say. Nothing good i mean.
Because:
- not really fast servers & network.
- the price of the nodes is overpriced.
- at this money i can find much better providers.
- i was pay for a brand, not for quality. There is a difference.
For now, after using them, i wont to use them anymore. I had no problems with servers, but it's just a trowing money for the wind.
They've always been considered the go-to for the high end VPS market. With these price cuts, their servers are the same price as DO, and 100% of the time I will choose Linode over DO.
There is much more than price to decide on. That is the issue, some people like one provider more, even if it costs more because of the other reasons. I am not saying DO is better than Linode, far from it, just that price is not all that is to be considering when you choose a host.
reviews & reputation much important for ISP then price.
Good support
Good host
No lags, troubles, downtimes
Why not to pay extra more $ for that?