Subject: Processors | February 7, 2018 - 09:01 AM | Tim Verry
Tagged: Xeon D, xeon, servers, networking, micro server, Intel, edge computing, augmented reality, ai
Intel announced a major refresh of its Xeon D System on a Chip processors aimed at high density servers that bring the power of the datacenter as close to end user devices and sensors as possible to reduce TCO and application latency. The new Xeon D 2100-series SoCs are built on Intel’s 14nm process technology and feature the company’s new mesh architecture (gone are the days of the ring bus). According to Intel the new chips are squarely aimed at “edge computing” and offer up 2.9-times the network performance, 2.8-times the storage performance, and 1.6-times the compute performance of the previous generation Xeon D-1500 series.
Intel has managed to pack up to 18 Skylake-based processing cores, Quick Assist Technology co-processing (for things like hardware accelerated encryption/decryption), four DDR4 memory channels addressing up to 512 GB of DDR4 2666 MHz ECC RDIMMs, four Intel 10 Gigabit Ethernet controllers, 32 lanes of PCI-E 3.0, and 20 lanes of flexible high speed I/O that includes up to 14 lanes of SATA 3.0, four USB 3.0 ports, or 20 lanes of PCI-E. Of course, the SoCs support Intel’s Management Engine, hardware virtualization, HyperThreading, Turbo Boost 2.0, and AVX-512 instructions with 1 FMA (fuse-multiply-add) as well..
Suffice it to say, there is a lot going on here with these new chips which represent a big step up in capabilities (and TDPs) further bridging the gap between the Xeon E3 v5 family and Xeon E5 family and the new Xeon Scalable Processors. Xeon D is aimed at datacenters where power and space are limited and while the soldered SoCs are single socket (1P) setups, high density is achieved by filling racks with as many single processor Mini ITX boards as possible. Xeon D does not quite match the per-core clockspeeds of the “proper” Xeons but has significantly more cores than Xeon E3 and much lower TDPs and cost than Xeon E5. It’s many lower clocked and lower power cores excel at burstable tasks such as serving up websites where many threads may be generated and maintained for long periods of time but not need a lot of processing power and when new page requests do come in the cores are able to turbo boost to meet demand. For example, Facebook is using Xeon D processors to serve up its front end websites in its Yosemite OpenRack servers where each server rack holds 192 Xeon D 1540 SoCs (four Xeon D boards per 1U sleds) for 1,536 Broadwell cores. Other applications include edge routers, network security appliances, self-driving vehicles, and augmented reality processing clusters. The autonomous vehicles use case is perhaps the best example of just what the heck edge computing is. Rather than fighting the laws of physics to transfer sensor data back to a datacenter for processing to be sent back to the car to in time for it to safely act on the processed information, the idea of edge computing is to bring most of the processing, networking, and storage power as close as possible to both the input sensors and the device (and human) that relies on accurate and timely data to make decisions.
As far as specifications, Intel’s new Xeon D lineup includes 14 processor models broken up into three main categories. The Edge Server and Cloud SKUs include eight, twelve, and eighteen core options with TDPs ranging from 65W to 90W. Interestingly, the 18 core Xeon D does not feature the integrated 10 GbE networking the lower end models have though it supports higher DDR4 memory frequencies. The two remaining classes of Xeon D SoCs are “Network Edge and Storage” and “Integrated Intel Quick Assist Technology” SKUs. These are roughly similar with two eight core, one 12 core, and one 16 core processor (the former also has a quad core that isn’t present in the latter category) though there is a big differentiator in clockspeeds. It seems customers will have to choose between core clockspeeds or Quick Assist acceleration (up to 100 Gbps) as the chips that do have QAT are clocked much lower than the chips without the co-processor hardware which makes sense because they have similar TDPs so clocks needed to be sacrificed to maintain the same core count. Thanks to the updated architecture, Intel is encroaching a bit on the per-core clockspeeds of the Xeon E3 and Xeon E5s though when turbo boost comes into play the Xeon Ds can’t compete.
The flagship Xeon D 2191 offers up two more cores (four additional threads) versus the previous Broadwell-based flagship Xeon D 1577 as well as higher clockspeeds at 1.6 GHz base versus 1.3 GHz and 2.2 GHz turbo versus 2.1 GHz turbo. The Xeon D 2191 does lack the integrated networking though. Looking at the two 16 core refreshed Xeon Ds compared to the 16 core Xeon D 1577, Intel has managed to increase clocks significantly (up to 2.2 GHz base and 3.0 GHz boost versus 1.3 GHz base and 2.10 GHz boost), double the number of memory channels and network controllers, and increase the maximum amount of memory from 128 GB to 512 GB. All those increases did come at the cost of TDP though which went from 45W to 100W.
Xeon D has always been an interesting platform both for enthusiasts running VM labs and home servers and big data enterprise clients building and serving up the 'next big thing' built on the astonishing amounts of data people create and consume on a daily basis. (Intel estimates a single self driving car would generate as much as 4TB of data per day while the average person in 2020 will generate 1.5 GB of data per day and VR recordings such as NFL True View will generate up to 3TB a minute!) With Intel ramping up both the core count, per-core performance, and I/O the platform is starting to not only bridge the gap between single socket Xeon E3 and dual socket Xeon E5 but to claim a place of its own in the fast-growing server market.
I am looking forward to seeing how Intel's partners and the enthusiast community take advantage of the new chips and what new projects they will enable. It is also going to be interesting to see the responses from AMD (e.g. Snowy Owl and to a lesser extent Great Horned Owl at the low and niche ends as it has fewer CPU cores but a built in GPU) and the various ARM partners (Qualcomm Centriq, X-Gene, Ampere, ect.*) as they vie for this growth market space with higher powered SoC options in 2018 and beyond.
- New Intel Xeon D Broadwell Processors Aimed at Low Power, High Density Servers
- Intel Xeon Scalable Processor Launch - New Architecture, New Platform for Data Center
- Qualcomm Centriq 2400 Arm-based Server Processor Begins Commercial Shipment
- Today's bonus AMD rumour: Starship, Naples, Zeppelin and a flock of Owls
*Note that X-Gene and Ampere are both backed by the Carlyle Group now with MACOM having sold X-Gene to Project Denver Holdings and the ex-Intel employee led Ampere being backed by the Carlyle Group.
Subject: General Tech | January 13, 2018 - 10:27 AM | Tim Verry
Tagged: WPA3, wifi alliance, wifi, wi-fi, networking, encryption
The Wi-Fi Alliance has announced an update to its Wi-Fi Protected Access (WPA) security suite in the form of WPA3. The first major update in more than a decade, WPA3 is a very welcome and much needed refresh with four new features aimed at both personal and enterprise networks.
Image courtesy Blue Coat Photos via Flickr Creative Commons.
The standards body did not go into many details on the new security suite, but did tease a few upcoming features in addition to closing known security vulnerabilities like KRACK. WPA3 uses a new 192-bit security suite "aligned with the Commercial National Security Algorithm (CNSA) suite from the Committee on National Security Systems" which is a collection of encryption techniques and algorithms that are reportedly up to the task of maintaining confidentiality on personal, enterprise, and industrial networks. Open Wi-Fi networks in particular will get the biggest boost from moving to WPA3 with support for individualized data encryption so that communication channels between the access point and users' devices are secured on a per-device basis. Personal networks also get improved security in the form of protections to protect users against themselves and maintain strong encryption even when they choose weak passwords. Setting up these security configurations is also being considered, and the Wi-Fi Alliance is promising easier configuration on devices with limited or no displays.
I am looking forward to more information on WPA3 as an update to WPA2 has been a long time coming. WEP has long been a joke and WPA2 has been vulnerable for a while so I hope that WPA3 lives up to its promises! What is not clear from the announcement is that if new hardware will be required or if WPA3 could be implemented through firmware and software updates. End user devices may be trickier to get updates from manufacturers, but perhaps wireless routers and access points can be upgraded without needing to buy new hardware. I suppose it depends on if radio and other hardware like the hardware accelerators / co processors need upgraded to support the new algorithms or not. In any case if you have been eyeing a new Wi-Fi AP or wireless router, maybe hold off for a few months to see how this shakes out.
Stay tuned for more information as it develops. What are your thoughts on WPA3 and the Wi-Fi Alliance's promises?
Subject: Networking | January 9, 2018 - 11:44 PM | Sebastian Peak
Tagged: Wireless-AC 1550, wireless, wi-fi, Rivet Networks, networking, killer, Intel, CES 2018, CES, 802.11ac Wave 2
For their new wireless adapter Rivet Networks has partnered with Intel, producing "the worlds fastest 2x2 11ac wireless networking adapter" in the Killer Wireless-AC 1550. This new adapter supports the 802.11ac Wave 2 standard and offers up to 1.73 Gbps throughput using 160 MHz channels.
"The first product to come out of Rivet Networks’ new partnership with Intel, the Killer™ Wireless-AC 1550 is the world’s fastest 2x2 11ac wireless networking adapter. The Killer Wireless-AC 1550 has been designed to combine the speed, intelligence, and control of Killer Networking products with the power and performance of the latest Intel wireless chipset. Delivering faster than gigabit Ethernet speeds along with the gaming functionality that gamers love, the Killer 1550 is the ideal wireless networking product for competitive gamers and performance users who demand the most from their computers."
Killer Networking lists these features for the Killer Wireless-AC 1550:
- Gigabit Wi-Fi Speeds: The Killer Wireless-AC 1550, featuring 160 MHz channel support, has a theoretical max throughput speed of 1.73Gbps when connected to a router that supports 160 MHz channels. This is faster than gigabit Ethernet and twice the speed of standard 2x2 11ac products.
- MU-MIMO Support: Killer 1550 includes full MU-MIMO (Multi-User-Multiple Input and Multiple-Output) support, which dramatically increases network efficiency by working with a MU-MIMO enabled access point. MU-MIMO allows wireless access points to support multiple transmissions at the same time, versus a single transmission at a time like normal access points. This creates additional efficiencies that can provide up to 60% faster download speeds, lower latency, and a better overall connection.
- Transmit Beamforming Technology: Killer 1550 also has Transmit Beamforming technology, which allows the Killer Wireless-AC 1550 to share location information directly to your wireless access point so that the access point can better direct its signals to you. This creates stronger wireless signals at all ranges and faster data transfers.
- Complete 802.11ac functionality: Supports dual band (2.4 GHz and 5GHz), IEEE standards-based 802.11a/b/g/n/ac, and includes Bluetooth 5.0 connectivity.
While the big news might be the Intel co-developed hardware, as this is a Killer Networking product the software is a big component in the overall experience. Options configurable via the Killer Control Panel include Advanced Stream Detect 2.0 for automated traffic prioritization for games and streaming, along with Lag and Latency Reduction Technology and Killer DoubleShot Pro support.
The first devices with the new Killer Wireless-AC 1550 adapter are being released this month.
Subject: General Tech | December 14, 2017 - 01:49 PM | Jeremy Hellstrom
Tagged: AT%26T, direcTV, security, networking, linksys
To start off with the bad news, as is our wont, DirecTV kits have a rather serious code injection problem. A researcher was able access the root shell on the Linksys WVBR0-25 wireless video bridge in less than 30 seconds, once he had access to one of the devices that the bridge was streaming to. As there are many infected machines out there, often PC's used only as video players as simple, poorly secured machines, this would mean your machines could be recruited into a botnet or mining pool quite easily. The researcher passed on his research to AT&T and Linksys 181 days ago he is quite disappointed they have yet to start develop a patch, according to The Register.
On a more positive note, AT&T is testing broadband over powerlines in Georgia and an undisclosed location outside the USA. They did not release any specifics of the current bandwidth which they can provide, though their goal is to surpass 1 gigabit per second. This will be quite the project as the testing we have done with powerline adapters did not show network connectivity anywhere near that speed in the best case scenarios, let alone when less than perfect wiring nor distance degraded the overall performance. You can check out more on that topic over at Slashdot.
"AT&T's DirecTV wireless kit has an embarrassing vulnerability in its firmware that can be trivially exploited by miscreants and malware to install hidden backdoors on the home network equipment, according to a security researcher."
Here is some more Tech News from around the web:
- How fast is a piece of string? Boffin shoots ADSL signal down twine @ The Register
- Crytek sues Star Citizen developers over game engine @ Ars Technica
- Flash bang walloped: Toshiba, Western Digital sign peace treaty over memory chip fabs @ The Register
- The Last Mile to Civilization 2.0: Technologies From Our Not Too Distant Future @ Techspot
- Intel to slap hardware lock on Management Engine code to thwart downgrade attacks @ The Register
- TSMC to spend $20bn on 3-nanometer chips @ Nikkei
- New battery boffinry could 'triple range' of electric vehicles @ The Register
Subject: General Tech | November 21, 2017 - 02:03 PM | Jeremy Hellstrom
Tagged: isp, networking, Internet, net neutrality, Autonomous System
If you are reading this from the US you probably have an opinion about the news out of the FCC today and should probably express that opinion to your various congress critters, even though Ajit Pai has stated he won't listen. As a backup plan you might want to take a read through this article over at Hack a Day which describes how you can set yourself up as your own ISP, aka an Autonomous System. The process is nowhere near as simple as setting up a home internet connection and you will need some dedicated equipment you may or may not have lying around. Those who live outside the USA should still take a look as there is some very interesting learning material in the article.
"It was during the purchase of data centre rack space that [Kenneth]’s challenge was laid down by a friend. Rather then simply rely on the connection provided by the data centre, they would instead rely on forging their own connection to the ‘net, essentially becoming their own Internet Service Provider."
Here is some more Tech News from around the web:
- Foldable Samsung Galaxy X smartphone support page leak suggests imminent release @ The Inquirer
- Intel finds critical holes in secret Management Engine hidden in tons of desktop, server chipsets @ The Register
- Researchers claim 400 of the world's top 50,000 websites track user keystroke behaviour @ The Inquirer
- US lab to build ARM-based supercomputer @ eeNews
- iPhone X: Bargain! You've just bagged yourself a cheap AR device @ The Register
- Another UAV licence price hike? Commercial drone fliers rage over consultation @ The Register
Subject: Networking | November 7, 2017 - 10:00 PM | Jim Tanous
Tagged: wi-fi, vpn, ubiquiti, networking, mesh, Amplifi HD, amplifi
Earlier this year we took a look at the AmpliFi HD Home Wi-Fi System as part of our review of mesh wireless network devices. AmpliFi is the consumer-targeted brand of enterprise-focused Ubiquiti Networks, and while we preferred the eero Mesh Wi-Fi System in our initial look, the AmpliFi HD still offered great performance and some unique features. Today, AmpliFi is introducing a new member of its networking family called AmpliFi Teleport, a "plug-and-play" device that provides a secure connection to users' home networks from anywhere.
Essentially a zero-configuration hardware-based VPN, the Teleport is linked with a user's AmpliFi account, which automatically creates a secure connection to the user's AmpliFi HD Wi-Fi System at home. Users take the small (75.85mm x 43mm x 39mm) Teleport device with them on the road, plug it in and connect it to the public Wi-Fi or Ethernet, and then connect their personal devices to the Teleport.
This provides a secure connection for private Internet traffic, but also allows access to local resources on the home network, including NAS devices, file shares, and home automation products. AmpliFi also touts that this would allow users to view their local streaming content even in locations where it would otherwise be unavailable -- e.g., watching U.S. Netflix shows while overseas, or streaming your favorite sports team while in a city where the game is blacked out.
In addition to traveling, AmpliFi notes that those with multiple homes or a vacation cottage could also benefit from Teleport, as it would allow you to share the same network resources and media streaming access regardless of location. In any case, a device like Teleport is still reliant on the speed and quality of your home and remote Internet connections, so there may be cases where network speeds are so low that it makes the device useless. That, of course, is a factor that would plague any network-dependent service or device, so while it's not a mark against the Teleport, it's something to keep in mind.
Teleport's features, while incredibly useful, are of course familiar to those experienced with VPNs and other secure remote connection methods. In terms of overall functionality, the AmpliFi Teleport isn't offering anything new here. The benefit, therefore, is its simple setup and configuration. Users don't need to setup and run a VPN on their home hardware, subscribe to a third party VPN service, or know anything about encryption protocols, firewall configuration, or network tunneling. They simply need to plug the Teleport into power, follow the connection guide, and that's it -- they're up and running with a secure connection to their home network.
You'll pay for this convenience, however, as the Teleport isn't cheap. It's launching today on Kickstarter with "early bird" pricing of $199, which will get you the Teleport device and the required AmpliFi HD router. A second round of early purchasers will see that price increase to $229, while final pricing is $269. Again, that's just for the Teleport and the router. A kit including two AmpliFi mesh access points is $399. There's no word on standalone pricing for the Teleport device only for those who already have an AmpliFi mesh network at home.
Regardless of the package, once you have the hardware there's no extra cost or subscription fee to use the Teleport, so frequent travelers might find the system worth it when compared to some other subscription-based VPN services.
The AmpliFi Teleport is expected to ship to early purchasers in December. We don't have the hardware in hand yet for performance testing, but AmpliFi has promised to loan us review samples as the product gets closer to shipping. Check out the Teleport Kickstarter page and AmpliFi's website for more information.
Subject: General Tech | September 29, 2017 - 12:53 PM | Jeremy Hellstrom
Tagged: icann, bind, dns, ksk, networking, security
ICANN have had to delay their planned upgrade to the root key signing keys used by DNS thanks to between 5-8% of key validators lacking the new KSK key. If a validator only possess the 2010 key, they would no longer be able to resolve DNS properly and the vast majority of the internet would disappear for stuck on the old system. The Register points out that the problem will actually be much larger as ICANN assumed that everyone has updated to the newest version of BIND DNS database, and only scanned those validators using the newest version.
The reason for the update is to increase the length of the root KSK that DNS depends on, which will greatly increase the security of anyone surfing the net and to help move this forward ICANN will be publishing a list of those out of date validators in the hopes publicity will spur them to upgrade. As with IPv6, we will wait and see.
"A multi-year effort to update the internet's overall security has been put on hold just days before it was due to be introduced, over fears that as many as 60 million people could be forced offline."
Here is some more Tech News from around the web:
- Benchmarks Show Firefox 57 Quantum Doing Well, But Chrome Largely Winning @ Phoronix
- TSMC announces plan to build 3nm fab in Taiwan @ DigiTimes
- Microsoft continues Linux love-in by joining the Open Source Initiative @ The Inquirer
- Ignite Overview @ Microsoft
- Microsoft gives all staff a marked-up 'Employee Edition' of Satya Nadella's new book @ The Register
- ZorinOS Is a Great Linux Desktop For Any User @ Linux.com
- Patch alert! Easy-to-exploit flaw in Linux kernel rated 'high risk' @ The Register
- Air Force Gives 10-Year-Old Orbiting Satellite To Ham Radio Operators @ Slashdot
- Whole Foods hacked and credit card info bagged @ The Inquirer
- E-Win Flash Series Gaming Chair @ TechPowerUp
Subject: General Tech | August 22, 2017 - 12:00 AM | Sebastian Peak
Tagged: WRT32X, router, networking, linksys, Killer Prioritization Engine, Killer Networking, gaming, AC3200
Linksys has announced a router that they say is 'built purely for gaming' with the WRT32X, an AC3200 router with a 1.8 GHz dual-core processor and built-in Killer Prioritization Engine.
"The WRT32X takes gaming to the next level. The router built purely for gaming features AC3200 speed and the Killer Prioritization Engine. The Killer Prioritization Engine identifies, prioritizes and accelerates gaming network traffic above all other devices in your home to deliver a faster, superior gaming experience. The Killer-enabled WRT32X also synchronizes with Killer-enabled PCs to give gaming traffic the highest priority on your network. Turning the Killer Engine on protects from extreme lag spikes and reduces lag by 77%, delivering consistent and superior reaction time during intense gaming scenarios."
Linksys lists the features of the WRT32X as follows:
- 1.8 GHz CPU: Dual-Core promotes simultaneous high-speed data processing.
- Pro-grade Gigabit Ethernet Switch: Gigabit (10/100/1000) is 10X faster than Fast Ethernet.
- Dual-Band (2.4 + 5 GHz): N600 + AC2600 Mbps.
- Killer Prioritization Engine: The first router that prioritizes gaming.
- Advanced Security: WPA2 encryption and SPI rewall help keep your network safely connected.
- Customized Gaming Interface: Custom-built interface and firmware for gaming traffic control.
- 256MB Flash and 512MB of RAM Memory: Handle more without delay for optimal performance.
- 4 High-Performance Antennas: Engineered to enhance dual-band communication; four external, adjustable antennas ensure supreme Wi-Fi signal strength.
- eSATA, USB 3.0, and USB 2.0 Ports: Share content via an external storage device with ultra-fast data transfer speeds. USB 3.0 delivers enhanced performance over USB 2.0; eSATA delivers optimal data transfer speeds from external SATA drives and accommodates USB 2.0.
The WRT32X carries an MSRP of $329.99, with availability TBA.
Subject: General Tech | May 28, 2017 - 07:10 PM | Tim Verry
Tagged: samba, linux, ransomware, security, networking
Last week, the development team behind Samba – popular software suite used on Linux and Unix clients and servers that uses TCP/IP protocol for file and print sharing to SMB/CIFS clients (including Microsoft Windows) – released a security advisory along with patches for a remote code execution hole that has been present in Samba for seven years since the release of Samba 3.5.0 in March 2010. The vulnerability, classified under CVE-2017-7494, allows an attacker to upload malicious code to a Samba server and get the server to run the code by sending a malformed IPC request that references the local file path. The Samba server will run the code in the malicious shared library (.so) file even though it is from an untrusted remote source.
The bad news is that this is a fairly serious flaw that could lead to an attacker successfully holding a business or home user’s files (including backups!) at ransom, stealing data, or using the now owned file server to attack other network resources that trust the file server. If not securely configured (e.g. allowing anonymous writes), the attack could even be wormable which would allow it to self-replicate across the network or Internet. Further, while various security firms have slightly different numbers, they all seem to agree that around 100,000 Internet-accessible machines are running vulnerable versions of Samba.
It is not all bad news though, and in some respects this vulnerability is not as big of an issue as the WannaCry ransomware and EternalBlue SMB vulnerability because in order to successfully exploit the Samba flaw an attacker needs to obtain credentials to upload the malicious code to the file share(s) which need to be writeable in the first place and not running as noexec under a SELinux policy. Also, attackers need to know or guess the local path name of the files on the file share to send the malformed IPC request. More importantly, the Samba team released three security releases (4.6.4, 4.5.10, and 4.4.14) for the newer branches and is working with OS distributions on providing patches for older Samba versions. For systems that cannot be updated or patched, there is also a workaround that can be implemented by modifying the global Samba config file to contain the setting “nt pipe support = no”. While this will break some expected Windows functionality (mainly machines will not be able to access null shares and will need to use the specific share path rather than just the server path), it will make it so that Samba will not accept the malicious requests.
Perhaps the most worrying aspect of this vulnerability is that security researchers estimate that up to 90% of the vulnerable Internet-connected Samba endpoints do not have a direct patch or update available yet and may not ever get one. While the enterprise hardware and even bigger consumer and SMB hardware providers will provide support for this in the form of patches or firmware updates, there is a sea of home routers, NAS boxes, file and print servers, and IoT devices running on home networks that are not open to user updates and may not ever get firmware updates. The best thing to do in this scenario according to the security advisory (if you can’t just not use it or replace it with different hardware that can be patched or isn’t affected of course) is to not expose it to the Internet. There would still be a risk of it being exploited should someone get a virus on a client machine through email, malicious downloads, or social engineering though. Considering these home NAS devices are usually used as destinations for backups, the risk of ransomware not only infecting client machines but also the main file share and network backups is scary. I have always been a fan of offline and/or cloud backups and in these modern times they are more important than ever with the rise of ransomware and other profit motivated viruses.
If you are not sure if your network is affected, there are tools being made available (including a Metasploit module, nmap scripts, and Internet scans) to help you determine that and reduce your attack surface using that information by updating to the latest security release, applying patches, updating, using SELinux policies to prevent the server from executing files itself, and preventing them from communicating with the Internet in order of effectiveness.
All that is to say don’t panic, stay vigilant, and make sure your important data is properly backed up and secured as much as possible!