Last night, in my graduate class in digital design, a man named John Crowley came in and gave about a 70 minute presentation on a product he was working on, and will soon begin to sell, called the Nimbus Cloud Computer. The reason why he was invited was mostly because he was a real-world entrepreneur who built a machine using FPGAs that had an Altera processor at their core (which is the same manufacturer as the processors for the FPGAs we use in the course).
The presentation was pretty much half sales pitch, half technical discussion. The Nimbus is a small machine, with no imbedded operating system, with only USB ports. It ran a virtual machine that connected to a Nimbus server. The cost of the box itself is about $50, but each user pays a monthly subscription fee per device. All of the “technical stuff” and troubleshooting is supposed to be handled by Nimbus customer service. I’m not entirely sure how that will work, but it’s an interesting concept.
The product has its obvious good and bad points.
Good: it’s fast, cheap, “green” (doesn’t draw a lot of power), easy to set up, accessible almost anywhere, and very secure.
Bad: video compression is very poor, and you’re limited to only using a few USB devices (pretty much a mouse and a monitor). He went on Youtube and played a video, and it was very, very choppy. It’s something they might be able to fix by porting over an MPEG to the board itself and decoding it there, but right now they’re doing a lot of compression up front, and it lags badly.
But Crowley brought up some interesting points. Since it’s very fast (a lot of routines are done via Verilog in hardware as opposed to software) and secure (no embedded operating system), it could have potential uses in the military and industries where speed and security are important. I don’t know if he mentioned it last night, but my co-worker Joe and I (who are both in the class) thought: why not hospitals?
I did a project for two months during my junior year in college in Bangkok, Thailand. We worked at a state-run hospital (Lerdsin Hospital to be specific) with the Orthopedic Department. The goal was to help build an electronic database to replace their paper system. We got the job done (most of the technical work was done by one of my project partners who was a CS major), but while doing the project, we were constantly asked about other technologies the doctors could use to access patient data, like a PDA or a tablet PC. It felt like stuff of pipe dreams, but, after the fact, we thought it was feasible, so we left it in our report as “possible future developments” in the recommendations section.
Last night I was reminded of that experience when Crowley talked about the mobility and security. With the age of health care reform in full swing, and paper records going the way of the Dodo, I thought it would be far more prudent for the company to sell the Nimbus to hospitals. It would help save cost AND allow doctors to access data cheaply and almost anywhere (the website says you can access a Nimbus via any Android enabled device). I even went so far as to write Crowley an e-mail saying so.
I’m not entirely sure how well ScreenPC, Crowley’s company, will do with the device. A lot of the small consumer ideas they have could easily be done via a Netbook, which costs about $50 to $100 more and has no monthly subscription fee. I do hope they go after hospitals.
Best of luck, ScreenPC.
The presentation was pretty much half sales pitch, half technical discussion. The Nimbus is a small machine, with no imbedded operating system, with only USB ports. It ran a virtual machine that connected to a Nimbus server. The cost of the box itself is about $50, but each user pays a monthly subscription fee per device. All of the “technical stuff” and troubleshooting is supposed to be handled by Nimbus customer service. I’m not entirely sure how that will work, but it’s an interesting concept.
The product has its obvious good and bad points.
Good: it’s fast, cheap, “green” (doesn’t draw a lot of power), easy to set up, accessible almost anywhere, and very secure.
Bad: video compression is very poor, and you’re limited to only using a few USB devices (pretty much a mouse and a monitor). He went on Youtube and played a video, and it was very, very choppy. It’s something they might be able to fix by porting over an MPEG to the board itself and decoding it there, but right now they’re doing a lot of compression up front, and it lags badly.
But Crowley brought up some interesting points. Since it’s very fast (a lot of routines are done via Verilog in hardware as opposed to software) and secure (no embedded operating system), it could have potential uses in the military and industries where speed and security are important. I don’t know if he mentioned it last night, but my co-worker Joe and I (who are both in the class) thought: why not hospitals?
I did a project for two months during my junior year in college in Bangkok, Thailand. We worked at a state-run hospital (Lerdsin Hospital to be specific) with the Orthopedic Department. The goal was to help build an electronic database to replace their paper system. We got the job done (most of the technical work was done by one of my project partners who was a CS major), but while doing the project, we were constantly asked about other technologies the doctors could use to access patient data, like a PDA or a tablet PC. It felt like stuff of pipe dreams, but, after the fact, we thought it was feasible, so we left it in our report as “possible future developments” in the recommendations section.
Last night I was reminded of that experience when Crowley talked about the mobility and security. With the age of health care reform in full swing, and paper records going the way of the Dodo, I thought it would be far more prudent for the company to sell the Nimbus to hospitals. It would help save cost AND allow doctors to access data cheaply and almost anywhere (the website says you can access a Nimbus via any Android enabled device). I even went so far as to write Crowley an e-mail saying so.
I’m not entirely sure how well ScreenPC, Crowley’s company, will do with the device. A lot of the small consumer ideas they have could easily be done via a Netbook, which costs about $50 to $100 more and has no monthly subscription fee. I do hope they go after hospitals.
Best of luck, ScreenPC.
No comments:
Post a Comment