The Planet Blog

I live and breathe technology, just like many of our customers. In speaking with my colleagues, they encouraged me to share my thoughts on a few technologies that are important to the Web hosting business.

In particular, clustering is top of mind for me right now. We talk a lot about clustering. So what does it really mean? Clustering involves multiple computers working together so they can be viewed as one. You’ve probably heard of a database cluster or a clustered Web server farm. These are prime examples of taking multiple servers and using clustering to give you high availability or high-performance computing which provides scalability that’s typically achieved by load balancing servers within the cluster.

One thing common with clusters is that the servers are typically setup independently and then joined with the cluster. This means that the operating system and applications are installed, and then configured to work as part of the cluster. For most clustering implementations, a single application instance cannot use the resources of more than one physical machine.

If you want an application to run on three servers, it requires that you install the application on three nodes in a cluster. Then the cluster can be set-up for load balancing the application across the three nodes.

Another critical clustering factor that shouldn’t be overlooked is storage. Clusters typically share a common back-end storage system so that all nodes in the cluster are working with the same data. Whether it’s a SAN with Web application content with all the nodes performing simultaneous reads in a load balancing cluster, or a direct attached storage subsystem that the nodes of a high-availability cluster exclusively access, the common shared storage is an integral part. For example, products may include Oracle Real Application Cluster (RAC), Microsoft Windows Compute Cluster Server, Sun Cluster and RedHat Cluster Suite.

Another hot topic is grid computing, which has many similarities to clusters and is often referred to interchangeably. A grid offers the ability to add or remove resources — like servers — from the grid as needed. This is referred to as “computing on demand” or “cloud computing.”

One distinction is that applications on a grid can typically scale to more than one physical server. Another nice feature of grid computing is that each node can be commodity hardware, so we can leverage the servers we already offer which provides a value to both our customers and to The Planet.

A grid is also well suited for running separate applications in a utility computing environment. Imagine if you had a SaaS application, and it ran on a grid where you subscribed to a minimum set of resources like CPU GHz, storage GBs and network Mb/sec. The grid would dedicate the minimum required resources to your application, and make sure there are enough resources in reserve to meet any bursting or scaling that may be required. It could add more servers to your grid on demand and free them up when they are no longer needed.

So these are the type of technologies that we’re evaluating at The Planet. Stay tuned for more details.

- Bryce

A question that I often face - what IS a traceroute? Most individuals know that it represents the “hops” or routers along the path from a system’s IP to the destination IP entered. What’s most often misunderstood is the response time numbers printed for each hop. Some assume that it is the time any packet takes to make that leg of the journey. But, that just isn’t the case.

Those times actually measure the time elapsed from when the packet was sent to when a response was received. In today’s Internet, most routers have very strict limits on how many of these responses they can generate in any particular time period. So a lack of response isn’t indicative of the latency of that leg, also because the router may use a very different part of its “brain” to generate these control responses than it would to simply forward a normal packet the times may be misleading. This means that a router under modest load may respond with wildly different times, as its busy doing other “housework.”

So how do we know where the lag is coming from with a traceroute? Only end-to-end pings can really show latency or packet loss. But, certain patterns in a traceroute can help pinpoint a possible source.

One method of detection is a cliff-like increase in latency that builds from one hop forward. In other words, the traceroute suddenly has a steady jump in the return times of each hop from one spot to the next. Notice I didn’t say stars in the route. In today’s Internet landscape, stars don’t reflect the certain issue they once did. Certain providers have restrained the routers so much that they constantly throw stars.

If you see every hop after a line is throwing stars then that link may be losing packets. An end-to-end ping showing this loss is just about the only way to verify that for sure. As if all these rate limits weren’t enough to render our poor traceroutes meaningless, there’s another issue making it even more difficult.

The Internet often involves paths that get somewhere a different way than they get back. In other words the Internet is asymmetric. This asymmetry means that the packet you sent to the hop in the traceroute got to the router one way - through one set of providers, links, etc. - and the router’s response got back to you another way. This means a lack of response, the star, or latency could indicate an issue with either path. It also means that the cliff I spoke of earlier could mean that past that hop the return path has an issue not the actual route you see. Yes, traceroute can only show you the outbound path. This is a weakness in the technique it uses. Only the outbound path is visible to the tracing packets. This makes diagnosing or finding the offending hop difficult if it’s located in the return path.

So how do we find the issue? Anyone that’s opened a network-related ticket with The Planet knows we like to have traceroutes (to indicate the path), then pings of 100 or so, from the IP at The Planet seeing the problem to the IP on the Internet. And if possible, and we know this isn’t very easy, the same from the Internet side of the issue back to The Planet IP.

If we have this sort of information we can usually determine where the problem exists. Of course like taking your car to the shop many times these traces and pings don’t show the problem because it is intermittent. They are still useful and at least give us a baseline to work from when we’re looking into the issue. Intermittent issues can also be helped by reporting times of day the issue happens and/or if it is limited to certain IPs or servers.

So the next time someone tells you that a 380ms spike in hop 5 means that the router is overloaded, or that a star in line 10 shows that we’re losing packets, you might let them know that it’s never that simple in today’s Internet.

- Turbo

Despite having traveled a fair bit and having been born in India, I’ve never been to China. I’ve always really wanted to go. Heck, my wife and I even considered getting married at the Forbidden Gardens here in Texas. As cool as that place is, I have to believe that the real deal is much more impressive.

While no one ever really needs one, it seems to me that the 2008 Olympic Games in Beijing might just give me the perfect excuse to go. If going to Beijing by itself wouldn’t be cool enough, its even more interesting when you hear about the preparation the city is going through, whether that be far fetched attempts at weather modification to stamping out bad translations on street signs all the way to mass education on etiquette in preparation for a global flock of visitors.

As ridiculous as this might sound, the work of the Olympics organizing committee reminds me a lot about my work here at The Planet. In preparation for the Olympics, this group has analyzed and anticipated the needs of their visitors/customers, built out solutions to meet those needs and ultimately are in the process of building out a scalable infrastructure that can accommodate a glut of demand.

Now, you’ve heard Will and Jeff talk about meeting our customer needs through building scalable networks or having N+1 redundancy within our data centers. While I’m always interested in what those guys are up to, my job is really about identifying needs and helping build solutions, which may sound pretty unbelievable when I say that I work in Marketing. In any case, many of you may remember we did a customer satisfaction survey and followed up to get a better view of what products and services our customers. Remember the banners in the customer portal?

First of all, I want to thank the thousands of you that took time to respond. There were a lot of quick hits that we learned and responded to. For instance 49% of you said that you were concerned about bandwidth prices and overages and we launched a promotion on unmetered bandwidth, or that 70% of you rated increased storage capacity as Important or Very Important, which led us to release our Managed Dedicated SAN offering . We heard you loud and clear on a number of others products and services, but some of them will take a bit longer before we can rollout, so stay tuned.

In the meantime, I wish we could get a little weather modification here before the summer heat really sets in.

- Urvish

a night doesn’t go by that I don’t find myself thinking about the planet’s current and future customers … about their business needs … their suggestions and comments … and how they rely on the planet to supply their IT solutions … it’s one of the things that keeps me up at night … i’ve spent a lot of time in the past few months … talking to lots of folks and asking questions … anytime I could corner some of our employees, customers or industry folks … picking up bits and pieces on what we can do better …

i’m very lucky to have a great group of folks at the planet and incredible customers … all willing to talk to us … in fact, we feature a number of our customer success stories on our web site … dickson supply, bigstockphoto.com and auto lead exchange to name just a few … these companies come to the planet for help in managing their hosting operations … they don’t want to manage server farms or internet providers … and don’t want to run their own data centers and man them 24×7x365 or worry about negotiating the best hardware deal … they’ve asked the planet to take on that responsibility …

no matter what business you’re in … to serve your customers you need to talk to them and really understand their business needs … there’s an old, old, old sales saying, “to sell john brown what john brown buys, you must learn to see through john brown’s eyes,” … don’t remember the author but I picked it up during ibm sales training … a couple of weeks ago we had a company meeting and talked about some of our customers … I talked with employees about customer success stories … about how companies rely on us to run their business … which is why we’ve got to be at the top of our game … i talked about ownership and follow through … about customer satisfaction and how we all work to support their needs …

i wanted all my employees to hear from a customer first-hand so they can continue to learn what “johnny” needs … so i invited a customer to come and speak to the company and tell them first-hand about why they picked us and what they expect from the planet in regards to support and service … this customer also happened to be one of our newer accounts that’s moving their entire server inventory to the planet from two other providers … the speaker did a super job … at the end of the meeting I heard several folks talking about the impact he made and how it really crystallized our responsibility to customers … we’re going to do more talking and listening to our customers … I know we can never talk to all 22,000 of you, but we can try … over the next quarter we’ll be launching several programs toward this end …

we have a great group of employees at the planet … and we’re continuing to invest in them with training and tools … and along the way we’ll bring talent in to fill certain needs … speaking of which … the latest addition to the senior management team is steve reichert … steve joins us as vice president of information technology and operations … he’ll be running the IT organization, software development and operations of our data centers in both cities …

steve comes to the planet with more than 27 years of experience in running IT organizations and data centers … most recently from infocrossing in omaha … in college he did his fortran and cobol programming on a 029 key punch that he lugged around in his backpack … just kidding … he’s trading in the corn fields and moving to dallas, where he lived for more than 16 years earlier in his career … you can read more about him on our web site … we’re thrilled to have him join the team … we’re working really hard to supply you with great products, excellent service and great results … we know we have more to do … in fact we’ll never totally be satisfied because excellence in never truly achieved … the yard sticks keeps moving …

thanks for your notes … all the input helps me learn … and even sleep better at night … be careful … it’s a jungle out there.

- Doug