OpenSolaris VPS

We ran unixbench 5.1.2 on one of our servers. We wanted to see how it compared to some of the competition. The Linode and EC2 benchmarks were done back in December 2009, on the latest hardware that was provided. Our OpenSolaris benchmarks were done on a system we bought back in March 2009.

We'll start off with Linode 360 plan (4 CPU / 4 parallel):

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   29911700.2   2563.1
Double-Precision Whetstone                       55.0       7852.9   1427.8
Execl Throughput                                 43.0       5470.0   1272.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     315110.5    795.7
File Copy 256 bufsize 500 maxblocks            1655.0      82099.9    496.1
File Copy 4096 bufsize 8000 maxblocks          5800.0     866155.2   1493.4
Pipe Throughput                               12440.0    2053207.3   1650.5
Pipe-based Context Switching                   4000.0     237263.9    593.2
Process Creation                                126.0      10784.4    855.9
Shell Scripts (1 concurrent)                     42.4       9259.1   2183.7
Shell Scripts (8 concurrent)                      6.0       1539.9   2566.5
System Call Overhead                          15000.0    1768915.5   1179.3
                                                                   ========
System Benchmarks Index Score                                        1254.5

Now, Amazon's EC2 medium (2 CPU / 4 parallel):

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   24194215.0   2073.2
Double-Precision Whetstone                       55.0       8422.0   1531.3
Execl Throughput                                 43.0       2379.7    553.4
File Copy 1024 bufsize 2000 maxblocks          3960.0     142163.5    359.0
File Copy 256 bufsize 500 maxblocks            1655.0      36551.4    220.9
File Copy 4096 bufsize 8000 maxblocks          5800.0     421398.5    726.5
Pipe Throughput                               12440.0     239183.7    192.3
Pipe-based Context Switching                   4000.0      82291.4    205.7
Process Creation                                126.0       2974.6    236.1
Shell Scripts (1 concurrent)                     42.4       5357.8   1263.6
Shell Scripts (8 concurrent)                      6.0        737.7   1229.5
System Call Overhead                          15000.0    1467331.5    978.2
                                                                   ========
System Benchmarks Index Score                                         579.6

Finally, our OpenSolaris VPS servers with about 20-30% utilization. The first one is with 2 CPU / 2 parallel:

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   54106806.1   4636.4
Double-Precision Whetstone                       55.0       7594.6   1380.8
Execl Throughput                                 43.0       1849.9    430.2
File Copy 1024 bufsize 2000 maxblocks          3960.0     126718.2    320.0
File Copy 256 bufsize 500 maxblocks            1655.0      32468.1    196.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     378133.6    652.0
Pipe Throughput                               12440.0    2062538.7   1658.0
Pipe-based Context Switching                   4000.0     262259.5    655.6
Process Creation                                126.0       2888.7    229.3
Shell Scripts (1 concurrent)                     42.4       4155.4    980.0
Shell Scripts (8 concurrent)                      6.0        728.6   1214.4
System Call Overhead                          15000.0     930953.7    620.6
                                                                   ========
System Benchmarks Index Score                                         724.0

We close out with 4 CPU / 4 parallel test on the same server:

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0  101277172.7   8678.4
Double-Precision Whetstone                       55.0      15197.7   2763.2
Execl Throughput                                 43.0       2764.1    642.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     152318.4    384.6
File Copy 256 bufsize 500 maxblocks            1655.0      39096.9    236.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     526719.9    908.1
Pipe Throughput                               12440.0    4177112.1   3357.8
Pipe-based Context Switching                   4000.0     650505.1   1626.3
Process Creation                                126.0       4924.3    390.8
Shell Scripts (1 concurrent)                     42.4       5666.8   1336.5
Shell Scripts (8 concurrent)                      6.0        806.4   1344.0
System Call Overhead                          15000.0     728384.9    485.6
                                                                   ========
System Benchmarks Index Score                                        1074.5

Clearly, we were not leaders in the benchmarks. But there is something interesting to be said about these results. Our OpenSolaris VPS servers make use of ZFS. ZFS does some interesting things. The File Copy tests lag behind greatly for one big reason. ZFS on our OpenSolaris VPS ensures reliability over performance. Let us explain.

Disks have cache on-board to help with speed. Most other operating systems and file systems will "tell" the application that a write is complete soon after the data is written to the disk cache (a temporary location, before it is written to the slower disk spindles). If the power were to fail at this point, the data is lost. Your business critical database thinks it wrote data to disk but in reality, it did not. It wrote to the cache. This leads to corruption and bad data.

ZFS doesn't do this. ZFS ensures that all data is properly flushed to disk on each commit. ZFS takes the safer and reliable approach to disk management. You can expect your database to be in good order if the system were to crash or if we were to ever experience a catastrophic power outage.

If we were to put aside the File Copy tests, we think we would have a pretty good chance of beating out the competition! In the future we'll run these same tests on a ZFS system with flush on commit turned off. But don't count on that being disabled on a production system. We prefer reliability over performance. Adding SSD to the mix should also improve things greatly, but you'll have to wait a little longer to experience OpenSolaris VPS with SSD flash disks.

Our main network provider is Internap, on the 10th floor in the Market Post Tower (San Jose) from CoreSite. We've recently started gathering quotes for a 2nd network provider. We came across Peer1. The negotiations fell through for one main reason:

If we added Peer1 to our mix, we would have to cross connect from 10th floor to Peer1's 16th floor. We weren't interested in dropping Internap and going with Peer1 only, so a cross connect was a must. There was a small catch (or rather a big catch): It would cost $2662 to put in a cross connect to our existing network provider in the same building. WHY was this so expensive!?

Market Post Tower is the most connect building, allowing providers to peer with other providers with ease. However, Peer1 not allowing us to do this was a big drawback (without paying lots of money). We dug a little deeper.

It turns out Peer1 has an office space rented from CoreSite which it converted to data center space. Peer1 has an isolated infrastructure from the rest of the building. Peer1 does not "peer" with CoreSite's Any2 network, so a cross connect would involve some expensive conduit installs. This is where the $2662 quote comes in.

There was no way for us to move our servers to Peer1 and still be connected to our current provider. If we had gone with Peer1, and down the line, if we wanted to not use Peer1 any longer, it would've involved some expensive cross connect charges.

Neither of which sounded interesting, so in the end we dropped Peer1 idea. We've decided to continue to use our premium "pure" Internap network inside Entic.net. We also didn't want to go from the "most connected building" status to a "least connected" floor status. :)

Our OpenSolaris VPS service is a niche market. Majority of the market is based on Linux and Xen (or other virtualization). Entic.net runs OpenSolaris whole root zones, nothing else. Here are some of the reasons why you might want to consider an OpenSolaris VPS from Entic.net and why WE are offering this service.

1. OpenSolaris non-global zones share a single kernel in the host OS ("global zone"). Your VPS talks directly to the kernel, and the kernel to the hardware. There is no other middle virtualization layer, improving performance. There are no bottlenecks. It is as if running a single OS on a single piece of hardware.

2. The single kernel also means that we get to manage the kernel level software, when we do upgrades. You don't have to worry about security issues in the kernel. The user level - you get a choice, either we can manage it or leave it to you. This is why we sometimes call our OpenSolaris VPS servers semi-managed.

3. OpenSolaris kernel is based on Solaris and Solaris is one of the oldest Unix variants out there. This means a rock solid foundation for the OpenSolaris VPS zones.

4. Our hardware and software are from the same company, Sun (and now, Oracle). We believe Sun has made several advancements in the OS to work very well with the Sun hardware. Likewise, the hardware is tuned well to be run on Solaris and OpenSolaris. Think about this: The hardware and software talk to each other through drivers. These drivers (in part) are made by the same company. Linux is quite the opposite. There is clear coordination, in the end, resulting in superior performance and reliability in the OpenSolaris VPS we offer.

5. OpenSolaris is a modern OS. It takes the kernel, which is very advanced with a long history and mixes it with some new user land technologies that have been invented with OpenSolaris in the past few year(s). These newer technologies are iterations of "what's best" out in the Linux community.

6. The most advanced file system out there runs on our OpenSolaris VPS. We all know this because literally all other companies out there are copying it. This of course, is ZFS. ZFS is what enables us to offer FREE disk space on our OpenSolaris VPS plans. It also makes it so you have very reliable service and allows for easy backups and restores (through snapshots).

There you have it. A quick run down of all the benefits. This is why we firmly believe that Entic.net's OpenSolaris VPS offering is the best platform for all your web site and critical hosting needs. Reliability. Stability. Performance. This is why we continue to offer OpenSolaris VPS servers.

Today, we've had a fairly large DNS outage. As some of you might be aware of, we host our DNS over at EveryDNS.com. It was free (well, we donated, and have supported it's cause through advertising on their site). It has served fairly well for the last several years, with only a few occasional DoS attacks.

Our LDAP address book (LDAPBook), mail and web site services were down. No doubt there would've been some access issues to our OpenSolaris VPS servers. Some of our customer's websites who also use EveryDNS were down.

We've taken DNS for granted.

One of the main reasons we've outsourced DNS is that we did not want to manage yet another service in-house. This is one of those things better left outsourced, leaving us free to concentrate on our core competencies - LDAP and OpenSolaris VPS.

We're investigating our options. We are considering using DynDNS, but one thing bothers us. This note on their web site:

Important note: Custom DNS cannot function if nameservers other than our nsX.mydyndns.org servers are listed in the delegation. Having "foreign" nameservers listed will prevent Custom DNS from initially activating, and later adding these extraneous nameservers will cause the service to deactivate within two weeks of the change. Please ensure only our five nameservers are listed in the delegation.

We do have secondary DNS servers in-house, and using that would be out of the question, since it seems we'll be forced to use only DynDNS name servers. This also brings up a point of what happens if DynDNS is DDoS attacked, since we'll be locked with one provider. However, DynDNS seems much more active in the DNS community (more than EveryDNS seems to have been), and they have real active support, which EveryDNS lacked. The pricing also seem to be quite high, we have over 30 domains (of which 10 are critical to our business) with EveryDNS and $30/year is about $900. DynDNS did offer to host our EveryDNS domains for free, for one year. But, after?

It's time to pick a provider that is flexible, and possibly not so expensive. We haven't ruled out DynDNS, but we haven't also jumped into the boat yet. We've taken DNS for granted, when DNS should be the most important infrastructure service. We'll change that...

zerigo.com looks interesting.

HAPPY HOLIDAYS!

Use the code "promo09" when checking out with Google Checkout. If you would like to pay through PayPal, we can offer the same coupon, but please contact us.

The 25% off is for the life of the account, so don't miss out on this opportunity to try out Entic.net's very fast and reliable OpenSolaris VPS servers (with a 99.99% uptime). SAS drives, dual power supplies, ZFS, DDR3 RAM, Intel Quad core.

Offer expires 12/31/2009.