Usually, we have a hard time getting shoes on Katie. She throws an incredibly huge fit. Today, however, we decided to take her outside for a walk and she thoroughly enjoyed it, up until the very end of course. We’ve uploaded some photos to the gallery and a video for your enjoyment.
I just got back from spending three days at Cisco’s corporate headerquarters. I’ve spent a good portion of my career dealing with vendors. I’ve spent days in meetings, evaluating products, specing out hardware, leased lines, and other ancillary pieces for building out networks and datacenters, but rarely have I left a set of meetings so jazzed about the future.
The first day was spent in design meetings and demo sessions. We opened the day by outlining where we’re currently at with our network. Currently, it’s a fairly well designed network with quite a few well known warts, but for the most part it functions very well. We’re not really hitting any capacity issues anywhere, but for the first time we’re really trying to plan out 2 to 5 years in advance. We know our current backbone, which is OC12 at the core nationwide, won’t last us for probably more than the next 12 to 18 months. We know this, because every time we improve our network, we find new ways to utilize it and we blow away all our traffic projections. However, we know that while our current design is working really well, we don’t want to upgrade 10 core locations to OC48 since half the sites sitting on that core won’t need that kind of connectivity. So we spent the first part of the morning drawing up a new core architecture with fewer sites and ideas for making it more highly available, including using DPR and/or MPLS fast reroute to provide sub-50 millisecond or sub-100 millisecond failover time in the event of a network failure. I think we have a promising set of ideas to go into our meetings next month with Level 3, who we purchase the majority of our core connectivity from.
The second half of the day was spent in demos. We saw some interesting new products on the Unified Communications front, as well a demo of Telepresence. Cisco makes this point often, but it can’t be stressed enough. Telepresence cannot be compared to video conferencing. We have an existing video conference system, which Telepresence has inspired me to improve, but it’s clunky and awkward by comparison. Telepresence really makes you feel like you’re sitting across the room from the other participants. Our current video conference system is used maybe once a month nationwide, but I guarantee Telepresence would be booked 100% in our company. I’d like to try to make our existing video conference system more like Telepresence, because a lot of the ideas Telepresence puts forward in terms of controlling the environment, using high quality video, and directional audio for the partipants can be adapted to our existing system. It won’t be Telepresence, but it certainly would improve what we currently have.
We finished up the day with a demo of products Cisco is pushing for retail and a discussion about their Service Control Engine product. The retail demo inspired a discussion about using EVDO as a backhaul for our retail stores, which is so obvious as to be a real slap in the face that we’re not already testing it out. We already own that last mile connectivity, why are we continuing to pay our competitors for connectivity to our retail outlets? Also, the Service Control Engine discussion was fascinating. We went from discussing a product we didn’t even know we owned to talking about all the things it could do for us. If you hear about providers shaping traffic down for Peer to Peer traffic and dunning customers for high bandwidth utilization, this is the product that’s doing it. It can shape traffic at wire speeds including multiple gigabit connections. Really really impressive.
The next two days were spent in Cisco’s Customer Proof of Concept lab. We went there to prove that the 7600 platform which we’re currently basing our core network architecture on will scale to meet our demands over the next 2 to 5 years. As I expected it did. I also had an excellent discussion with a Cisco executive over in the Unified Communications area about how we can leverage our investment in our IP Telephony infrastructure to really improve communications and productivity.
Cisco really went out of their way to get us access to people who would could answer our questions, give us good ideas for the future and give us access to equipment to test our their claims for our own assurance. They’re really the gold standard that I’m holding everyone else up to in terms of how they should treat their customers. We could have spent the same, dollar for dollar, with many other vendors and barely gotten a sales engineer to give us a call back, but Cisco seems to understand from the top down that people sell products, the products don’t sell themselves. They understand that we’re looking for a vendor to partner with us to help us get to our next level. Most often in this business there are two or more products that will do any given job, and while technically one might be superior, in the end both would work for a given task. The decision comes down to more than what exactly the product does, and while having a superior product helps, having a superior organization that stands behind that product and offers it at competitive prices is in my book the best way to get your product into customers hands. Cisco has a product for nearly anything you need in the networking world, and while we won’t ever be exclusively a Cisco shop, they’re much more likely to sell us a product on a given day due to the fact that they spend so much time focusing on helping their customers implement their solutions and spend so much time focusing on helping their customers support their existing installations. At the end of the day, the most important question I have to ask when making a purchasing decision usually isn’t what is the best product, it’s which is the best company. Right now, Cisco’s going a long way to convincing me they are.
We recently purchased a Canon SD1000 for my 27th birthday. It’s an awesome, incredibly compact camera, unlike our old Nikon which was a total brick and never got used because it was inconvenient to tote around. I’m hoping we’ll be taking a lot more pictures since the camera will be a lot more convenient to have with us. In the mean time, we’ve added a gallery from today’s testing to show off the new camera. Keep checking back, Katie will be featured often. Here’s a sample image of the most beautiful girl in the world:
Once again, open source proves its usefulness. Currently we’re implementing a rather large network monitoring setup from several large companies, mostly with TLAs as their company names. These cost millions of dollars to buy, hundreds of thousands of dollars to customize and equally as much to get the basics working in your environment. I used to use a tool called tkined back in the day to monitor our ISP network via bandwidth graphs overlayed on a network diagram. This was real-time and was the ultimate in network monitoring at the time. Strangely, the tools have all advanced, but I had yet to find something to do an equivalant graphical view of the network. That was until I found PHP Weathermap, which can be installed as a Cacti plugin. We already had a small Cacti installation that just needed to be expanded to include the majority of our core routers (it was mainly being used for ancillary routers and switches), and then I drew the map on top of an existing Visio I exported to PNG. The result:
This gives a near real time view of network bandwidth utilization. It was exactly what I was looking for. When I discussed this with the team doing the implementation from the TLA company, they said it might be possible, but I have a feeling it’s going to cost more than it would take a programmer to rewrite this from the ground up. Thus is enterprise software: the more expensive the package, the less you get out of the box and the more likely you are to end of spending hundreds of thousands if not millions only to replace it with something simpler.
I obviously watch way more television than I should. For some reason these days I find that when I get home from a hard day’s work that I prefer passive entertainment to sitting down and writing my thoughts here. This should seem obvious to anyone who has noticed my posting frequency since I took my new job.
This season, I avidly watched Jericho on CBS. Jericho is a show about a post-apocalyptic society in a small town, called Jericho, in Kansas. Terrorists detonate nuclear bombs in most major cities across the US, and the inhabitants of Jericho must live with the effects of having a broken economy, lack of electricity, food, running water, and a total lack of information from the outside world (although in theory parts of the Internet would have still survived, although that’s I guess addressed by the lack of electricity). There are shortages of nearly everything needed to survive, including food and water, and to make matters worse, the neighboring town of New Bern is coveting what Jericho has and has declared war on Jericho to take large portions of the town away. This is where the season finale cliffhanger left us.
That is, that’s what we thought was the season finale. Apparantly CBS, in it’s infinite wisdom, decided that more reality TV and more cop drama were the answer to their ratings woes. They stuck Jericho up against American Idol, the number one show on television, and then blame the viewers for not tuning in when they decided to cancel the show. The die-hard fans, including myself, are fighting back by sending thousands of pounds of nuts to CBS headquarters (an inside reference to an integral part of the plot of the season finale). If you’ve watched the show, or even if you just care about having quality produced dramas rather than more reality TV drivel and cop dramas on prime-time television, please donate to the cause at the Nuts for Jericho site.
Powered by WordPress