Good uptime is great, but unfortunately very expensive in terms of hardware, software and manpower. Questions you should ask yourself: - What is the maximum allowable downtime duration ? - How many outages can you tolerate per year ? - What is actual cost to you of a one day/evening outage ? - How many such outages did you have with your actual infrastructure ? I think the best option in your case is to have two identical servers/PCs of good quality with two mirrored harddrives each in hot-swap slots. If a harddrive fails, you can carry on for the evening and replace it the next day. If something else fails you swap the SQL server drives into the second server/PCs and fix the problem later. This is simple enough that you can instruct someone by phone to do that, when you are absent yourself.

This looks to me to be similar to Bluegene supercomputers. A Bluegene essentially consists of packaged PowerPC processors with a scalable high-performance switch interface on board. The two first current generation Bluegenes were using 32bit CPUs as well.

Markus

This Google experiment proves that you can build a self driving car who can drive safely for thousands of miles on actual public roads. Yes, there are some additional conditions which are not practical (like the driving the path manually before to record a precise map), but it has been done.

Most drivers are not very much 'observing', I just read somewhere that the Blackberry outage caused vehicle accidents to drop by 20-40% in some Gulf states. Observing drivers, Ha !

Just compare how many accidents are caused by trees falling onto the street compared to texting.

The biggest problems to solve are the cultural, legal and liability issues.

Markus

I'm very much in favor of my car driving instead of me. I'm sure it will drive safer and better. For me the daily commute is a chore and I'd be very happy to leave it to a machine.

I'm sure there will be viruses/sabotage. But I'm also sure human drives cause more accidents than viruses/sabotage will.

Markus

I you are doing wrong, then. My Unix and Linux systems need rebooting only because we install a OS update about once a year.

Markus

> Even if you decided to maintain a VM system, the idea of a unified storage system (DRAM+DISK as one device) is pretty fascinating.

You can buy this from IBM since >20 years. It is called 'System/38', 'AS/400' or 'iSeries'. The system has a flat 64-bit address space and memory essentially serves as cache for disk.

Markus

Such projects are always based on estimated performance numbers. It looks to me like the estimated (and contractually signed) target performance was higher than what IBM could deliver in the budget envelope and the target timeframe. Probably the technology advance was not delivering as expected. As the U of Illinois was not ready to soften on some aspects to make it fit IBM had the choice of either delivering much more hardware at a loss or to pull out.

From my (limited) search it looks like the project was signed in 2008, so back then IBM estimated they can deliver a PetaFLOP for $200M in 2011. It looks like they were wrong.

I witnessed similar situation where the machine could deliver the promised performance using a benchmarking program, but real apps were unable to get to similar numbers. Improving the software stack made the performance available to apps, but it took 2 years to get there. The hardware was performing well, but immature software was spoiling it.

Markus

That is exactly what will not happen.

The ones who should tell their Customers about the problem is Siemens. But they will play the problem down because it might affect the sales of the next batch of stuff.

The evil hacker will just buy a bunch of systems, analyze it and find the vulnerabilities. This completely independent of the disclosure. Stuxnet was developed before this disclosure and I think the vulnerabilities used by Stuxnet are still there.

This is why security by obscurity does not work in the real world.

I alway think defending against such lasers is quite easy and low-cost: Just put up a good-quality mirror and the beam gets reflected. If you aim well, you could even attack the ship with its own laser-beam.

Markus

I would do battery compression.

Imagine: One core for the phone and three cores transparently decompressing the battery in the background. You'd get three times the battery life !

Finally a smartphone with a battery lasting a full week !

I don't know about SGI, but I do know that the biggest AIX/Power7 box you can get today has 256 cores (32 chips with 8 cores each). I know there was kernel work involved but the scaling problems I remember mentioned involved the number threads (1024) as the P7 core has 4-way multithreading.

The Linux certainly may need work for such machines, but a new operating system ? bullshit.

Use standard HTML for as much as possible. Complement the rest with flash.

If you choose Silverlight you'll exclude automatically all platforms which are not Windows mainstream (Vista and 7). Flash is well supported about everywhere.

I'm typing this on a Ubuntu workstation with Chrome. No Silverlight available here.

You don't post any specific minimal specs you need, so here what I'd want for my lab:

• Four channels, quite often two channels are a bit short
• Color, with four channels color helps a lot to identify the signals
• DSO, goes without mention
• Computer-interface, mostly for screen shots to document and share
• Built-in logic analyzer, many times you'll have a combination of analog and digital signals to watch

I've seen many intelligent discussion on avrfreaks.net about the topic Oscilloscope search on avrfreaks

Markus

I hope Motorola get's a nice class-action suit out of this.

Imagine a nice little virus, designed to trigger the 'self-destruct' and some innocent users getting infected.

Markus

I'd like that screen on my mobile phone. That's where I'd need a sunlight-readable, battery conserving display most. Most GPS functions only work outside due to feeble GPS signals, but at the same time the display become almost unreadable.

There are plenty of business opportunities and markets for Mary Lou to explore !

Markus