Disks from the Perspective of a File System

Most applications do not deal with disks directly, instead storing their data in files in a file system, which protects us from those scoundrel disks. After all, a key task of the file system is to ensure that the file system can always be recovered to a consistent state after an unplanned system crash (for example, a power failure). While a good file system will be able to beat the disks into submission, the required effort can be great and the reduced performance annoying. This article examines the shortcuts that disks take and the hoops that file systems must jump through to get the desired reliability.

via Disks from the Perspective of a File System – ACM Queue.

Luckily, SATA (serial ATA) has a new definition called NCQ (Native Command Queueing) that has a bit in the write command that tells the drive if it should report completion when media has been written or when cache has been hit. If the driver correctly sets this bit, then the disk will display the correct behavior.

In the real world, many of the drives targeted to the desktop market do not implement the NCQ specification. To ensure reliability, the system must either disable the write cache on the disk or issue a cache-flush request after every metadata update, log update (for journaling file systems), or fsync system call. Both of these techniques lead to noticeable performance degradation, so they are often disabled, putting file systems at risk if the power fails. Systems for which both speed and reliability are important should not use ATA disks. Rather, they should use drives that implement Fibre Channel, SCSI, or SATA with support for NCQ.

On Linux here’s how you can check if your drive has NCQ.

$ cat /sys/block/sd?/device/queue_depth

A 1 indicates no NCQ.  and

$ cat /sys/block/sd?/device/queue_type

My green drives came back with none.

Market Leaders Drive Investment in Switching

Terabit switch chips and switch chipsets that scale up to 400 Tbit/s are making merchant switch chips very attractive for high-performance network systems against in-house ASIC and FPGA-based designs. The latest switch devices not only deliver on performance, but integrate additional system functions such as Ethernet MACs and programmable classification engines to support software-defined networking (SDN).

via Light Reading – Market Leaders Drive Investment in Switching – Telecom.

High-performance switch devices have become both more complicated and simpler. The devices integrate additional functionality with significant on-chip memory and packet processing functions. On the other hand, the interfaces for most protocols are moving to 10 Gbit/s for the current generation and 25 Gbit/s for the next generation.

Scientists Establish First Working Quantum Network

The team has managed to rig up a laser to fire and hit the first networked atom in a way that the atom preserves its quantum state, but also produces a photon with that information plastered onto it. The photon then shoots off down the fiber optic cable delivering it to the second atom. Network achieved. On top of that, the researchers managed to get the two networked atoms to entangle, which means the network should be completely scalable to something along the lines of an Internet.

via Scientists Establish First Working Quantum Network | Geekosystem.

NOVA series on quantum mechanics.

And another link from Cornell: Does quantum entanglement imply faster than light communication?

Say you agree to send out two beams of light to your two friends who live on opposite sides of the galaxy (you live in the middle). Ahead of time you tell them that if one of the beams of light is red the other will be blue. So you send the blue beam to your friend on one side and immediately she knows that your other friend is recieving a red beam at the same time. Aha! You say, my friends have now communicated at a speed faster than the speed of light and violated relativity, but no real information has been passed between them. You have told both of them at a normal sub-luminal speed about what you just did and that’s all. (A way of proving there’s no faster than light communication is that you could lie and send them both the same coloured beam of light and they would never know!).

Watching quantum mechanics in action: Researchers create world record laser pulse

UCF Professor Zenghu Chang from the Department of Physics and the College of Optics and Photonics, led the effort that generated a 67-attosecond pulse of extreme ultraviolet light. The results of his research are published online under Early Posting in the journal Optics Letters.

via Watching quantum mechanics in action: Researchers create world record laser pulse.

File and Data Storage: AFS

AFS Andrew File System is a distributed, networked file system that enables efficient file sharing between clients and servers. AFS files are accessible via the Web or through file transfer programs such as OpenAFS or Fetch Macintosh and SecureFX Windows. Currently all users with a full-service SUNet ID are granted 2 GB of AFS file space. Additional disk space is available by request for faculty-sponsored research including dissertations.

via File and Data Storage: AFS | Information Technology Services.

On the Feasibility of Side-Channel Attacks with Brain-Computer Interfaces

The security risks involved in using consumer-grade BCI devices have never been studied and the impact of malicious software with access to the device is unexplored. We take a first step in studying the security implications of such devices and demonstrate that this upcoming technology could be turned against users to reveal their private and secret information. We use inexpensive electroencephalography (EEG) based BCI devices to test the feasibility of simple, yet effective, attacks. The captured EEG signal could reveal the user’s private informa- tion about, e.g., bank cards, PIN numbers, area of living, the knowledge of the known persons. This is the first attempt to study the security implications of consumer-grade BCI devices. We show that the entropy of the private information is decreased on the average by approximately 15 % – 40 % compared to random guessing attacks.

via On the Feasibility of Side-Channel Attacks with Brain-Computer Interfaces | USENIX.

YouTube Flags Democrats’ Convention Video on Copyright Grounds

On Wednesday morning, a campaign spokesman confirmed there was a “technical error on YouTube that inadvertently triggered a copyright message at the end of the live stream Tuesday night,” adding “We do not expect tonight’s coverage will be affected.”

After this story was published, the video was subsequently marked “private.”

via YouTube Flags Democrats’ Convention Video on Copyright Grounds | Threat Level | Wired.com.

Intel immerses its servers in oil — and they like it!

The idea of immersing servers in oil to keep them cool isn’t entirely new — passionate gamers have been housing their systems in vegetable oil for years. But it’s time to take notice of this trend when Intel starts singing its praises as a potentially revolutionary method for slashing the price of running a data center.

via Intel immerses its servers in oil — and they like it! — Cloud Computing News.

Most servers today follow design principles for optimal airflow, but “we could throw some of those rules out,” he said, and maybe build a better server. The obvious steps are eliminating anything to do with fans, sealing hard drives (or going to solid-state drives) and replacing any organic materials that might leech into the oil. A redesign of the heat sink probably would be in order, as would a rethinking of where things sit on the motherboard.

TSMC plans 450mm wafers by 2018

The 450mm wafers would help solve the problem of rising costs in making advanced chips, allowing TSMC to provide affordable 10 nanometer chips with FinFET transistors for customers, J.K. Wang (王建光), vice president of TSMC’s operation in charge of 300mm factories, told a media briefing arranged by semiconductor industry association SEMI.

via TSMC plans 450mm wafers by 2018 – Taipei Times.

Chipmakers can get 2.5 times more chips from a 450mm wafer than from a 300mm wafer.