Habana Labs has unveiled the Gaudi HL-2000, a custom-built AI processor that the company claims can outrun Nvidia’s best and brightest GPUs at training neural networks. …
AI Chip Startup Releases Training Accelerator to Challenge GPUs was written by Michael Feldman at .
The startup’s earlier AI chip is called Goya, and Facebook uses this inference processor for its...
Big targets: Internet-connected security cameras make up nearly half of all the Internet of Things devices compromised by hackers, ZDNet reports. Smart hubs and network-attached storage devices are next on the hit list. The average U.S. household contains 17 Internet-connected devices, while the average European household has 14.
New rules for the IoT? A U.S. House of Representatives committee has approved an IoT security bill that would create security standards that vendors would have to apply before government agencies could buy IoT devices from them, Nextgov says. The legislation aims to leverage the government’s substantial purchasing power to drive security in the IoT market.
Cleaning house: The Chinese government has blocked several foreign media sites in the name of cleaning up the Internet, Reuters reports. China’s campaign will punish and expose websites for “illegal and criminal actions” and for failing to “fulfil their obligation” to take safety measures or prevent the theft of personal information, the government says.
Exposing anti-encryption: The Electronic Frontier Foundation and other groups are going to court in an attempt to expose the U.S. Department of Justice’s efforts to break Facebook’s Messenger encryption, ZDNet reports. The groups are asking a court to unseal Continue reading
To kick-off Crypto Week 2019, we are really excited to announce a new solution to a long-standing problem in cryptography. To get a better understanding of the technical side behind this problem, please refer to the next post for a deeper dive.
Everything from cryptography to big money lottery to quantum mechanics requires some form of randomness. But what exactly does it mean for a number to be randomly generated and where does the randomness come from?
Generating randomness dates back three thousand years, when the ancients rolled “the bones” to determine their fate. Think of lotteries-- seems simple, right? Everyone buys their tickets, chooses six numbers, and waits for an official to draw them randomly from a basket. Sounds like a foolproof solution. And then in 1980, the host of the Pennsylvania lottery drawing was busted for using weighted balls to choose the winning number. This lesson, along with the need of other complex systems for generating random numbers spurred the creation of random number generators.
Just like a lottery game selects random numbers unpredictably, a random number generator is a device or software responsible for generating sequences of numbers in an unpredictable manner. As the need for Continue reading
Randomness, randomness everywhere;
Nor any verifiable entropy.
Generating random outcomes is an essential part of everyday life; from lottery drawings and constructing competitions, to performing deep cryptographic computations. To use randomness, we must have some way to 'sample' it. This requires interpreting some natural phenomenon (such as a fair dice roll) as an event that generates some random output. From a computing perspective, we interpret random outputs as bytes that we can then use in algorithms (such as drawing a lottery) to achieve the functionality that we want.
The sampling of randomness securely and efficiently is a critical component of all modern computing systems. For example, nearly all public-key cryptography relies on the fact that algorithms can be seeded with bytes generated from genuinely random outcomes.
In scientific experiments, a random sampling of results is necessary to ensure that data collection measurements are not skewed. Until now, generating random outputs in a way that we can verify that they are indeed random has been very difficult; typically involving taking a variety of statistical measurements.
During Crypto week, Cloudflare is releasing a new public randomness beacon as part of the launch of the League of Entropy. The League of Entropy is Continue reading
Creating the Tesla GPU compute platform has taken Nvidia the better part of a decade and a half, and it has culminated in a software stack comprised of various HPC and AI frameworks, the CUDA parallel programming environment, compilers from Nvidia’s PGI division and their OpenACC extensions as well as open source GCC compilers, and various other tools that together account for tens of millions of lines of code and tens of thousands of individual APIs. …
Nvidia Makes Arm A Peer To X86 And Power For GPU Acceleration was written by Timothy Prickett Morgan at .
One of the first things I realized when I started my Azure journey was that the Azure orchestration system is incredibly slow. For example, it takes almost 40 seconds to display six routes from per-VNIC routing table. Imagine trying to troubleshoot a problem and having to cope with 30-second delay on every single SHOW command. Cisco IGS/R was faster than that.
If you’re old enough you might remember working with VT100 terminals (or an equivalent) connected to 300 baud modems… where typing too fast risked getting the output out-of-sync resulting in painful screen repaints (here’s an exercise for the youngsters: how long does it take to redraw an 80x24 character screen over a 300 bps connection?). That’s exactly how I felt using Azure CLI - the slow responses I was getting were severely hampering my productivity.
Read more ...If you want to rank in the world of supercomputing – as national and state governments as well as academic and research institutions most certainly do to prove their worth for research and development – then you need to have at least a petaflops ante now to play the game. …
Supercomputing Coiled To Spring To Exascale was written by Timothy Prickett Morgan at .
Towards multiverse databases Marzoev et al., HotOS’19
A typical backing store for a web application contains data for many users. The application makes queries on behalf of an authenticated user, but it is up to the application itself to make sure that the user only sees data they are entitled to see.
Any frontend can access the whole store, regardless of the application user consuming the results. Therefore, frontend code is responsible for permission checks and privacy-preserving transformations that protect user’s data. This is dangerous and error-prone, and has caused many real-world bugs… the trusted computing base (TCB) effectively includes the entire application.
The central idea behind multiverse databases is to push the data access and privacy rules into the database itself. The database takes on responsibility for authorization and transformation, and the application retains responsibility only for authentication and correct delegation of the authenticated principal on a database call. Such a design rules out an entire class of application errors, protecting private data from accidentally leaking.
It would be safer and easier to specify and transparently enforce access policies once, at the shared backend store interface. Although state-of-the-are databases have security features designed for exactly this purpose, Continue reading
The Internet is an extraordinarily complex and evolving ecosystem. Its constituent protocols range from the ancient and archaic (hello FTP) to the modern and sleek (meet WireGuard), with a fair bit of everything in between. This evolution is ongoing, and as one of the most connected networks on the Internet, Cloudflare has a duty to be a good steward of this ecosystem. We take this responsibility to heart: Cloudflare’s mission is to help build a better Internet. In this spirit, we are very proud to announce Crypto Week 2019.
Every day this week we’ll announce a new project or service that uses modern cryptography to build a more secure, trustworthy Internet. Everything we release this week will be free and immediately useful. This blog is a fun exploration of the themes of the week.
Many pieces of the Internet in use today were designed in a different era with different assumptions. The Internet’s success is based on strong foundations that support constant reassessment and improvement. Sometimes these improvements require deploying new protocols.
Performing an upgrade on a system Continue reading
Cloudflare believes trust is fundamental to helping build a better Internet. One way Cloudflare is helping our customers earn their users’ trust is through industry standard security compliance certifications and regulations.
Security compliance certifications are reports created by independent, third-party auditors that validate and document a company’s commitment to security. These external auditors will conduct a rigorous review of a company’s technical environment and evaluate whether or not there are thorough controls - or safeguards - in place to protect the security, confidentiality, and availability of information stored and processed in the environment. SOC 2 was established by the American Institute of CPAs and is important to many of our U.S. companies, as it is a standardized set of requirements a company must meet in order to comply. Additionally, PCI and ISO 27001 are international standards. Cloudflare cares about achieving certifications because our adherence to these standards creates confidence to customers across the globe that we are committed to security. So, the Security team has been hard at work obtaining these meaningful compliance certifications.
Since the beginning of this year, we have been renewing our PCI DSS certification in February, achieving SOC 2 Type 1 compliance in March, obtaining Continue reading
Argo Tunnel lets you expose a server to the Internet without opening any ports. The service runs a lightweight process on your server that creates outbound tunnels to the Cloudflare network. Instead of managing DNS, network, and firewall complexity, Argo Tunnel helps administrators serve traffic from their origin through Cloudflare with a single command.
We built Argo Tunnel to remove the burden of securing and connecting servers to the Internet. This new model makes it easier to run a service in multi-cloud and hybrid deployments by replacing manual and error-prone work with a process that adds intelligence to the last-mile between Cloudflare and your origins or clusters. However, the service was previously only available to users with Cloudflare accounts. We want to make Argo Tunnel more accessible for any project.
Starting today, any user, even those without a Cloudflare account, can try this new method of connecting their server to the Internet. Argo Tunnel can now be used in a free model that will create a new URL, known only to you, that will proxy traffic to your server. We’re excited to make connecting a server to the Internet more accessible for everyone.
Argo Tunnel replaces Continue reading
The Packet Pushers' latest crash course podcast on wireless for wired engineers dives into beacons, roaming, SSIDs, the ugly reality of band steering, and more WLAN nerdery. Our guest is Chris Reed.
The post Heavy Networking 455: The Wonderful World Of Wireless – Beacons, Probes, SSIDs And More appeared first on Packet Pushers.
Over the past two years Docker has worked closely with customers to modernize portfolios of traditional applications with Docker container technology and Docker Enterprise, the industry-leading container platform. Such applications are typically monolithic in nature, run atop older operating systems such as Windows Server 2008 or Windows Server 2003, and are difficult to transition from on-premises data centers to the public cloud.
The Docker platform alleviates each of these pain points by decoupling an application from a particular operating system, enabling microservice architecture patterns, and fostering portability across on-premises, cloud, and hybrid environments.
As the Modernizing Traditional Applications (MTA) program has matured, Docker has invested in tooling and methodologies that accelerate the transition to containers and decrease the time necessary to experience value from the Docker Enterprise platform. From the initial application assessment process to running containerized applications on a cluster, Docker is committed to improving the experience for customers on the MTA journey.
Enterprises develop and maintain exhaustive portfolios of applications. Such apps come in a myriad of languages, frameworks, and architectures developed by both first and third party development teams. The first step in the containerization journey is to determine which applications are Continue reading
sudo mn --custom sflow-rt/extras/sflow.py --link tc,bw=10 \Mininet, ONOS, and segment routing provides an example using a Custom Topology, e.g.
--topo tree,depth=2,fanout=2
sudo env ONOS=10.0.0.73 mn --custom sr.py,sflow-rt/extras/sflow.py \This article describes how to incorporate sFlow monitoring in a fully custom Mininet script. Consider the following simpletest.py script based on Working with Mininet:
--link tc,bw=10 --topo=sr '--controller=remote,ip=$ONOS,port=6653'
#!/usr/bin/python
from mininet.topo import Topo
from mininet.net import Mininet
from mininet.util import dumpNodeConnections
from mininet.log import setLogLevel
class SingleSwitchTopo(Topo):
"Single switch connected to n hosts."
def build(self, n=2):
switch = self.addSwitch('s1')
# Python's range(N) generates 0..N-1
for h in range(n):
host = self.addHost('h%s' % (h + 1))
self.addLink(host, switch)
def simpleTest():
"Create and test a simple network"
topo = SingleSwitchTopo(n=4)
net = Mininet(topo)
net.start()
print "Dumping host connections"
dumpNodeConnections(net.hosts)
print "Testing bandwidth between h1 and h4"
h1, h4 = net.get( 'h1', 'h4' )
net.iperf( (h1, h4) )
net.stop()
if __name__ == '__main__':
# Continue reading
Some analysts now think the deal's approval chances are at less than 50%.
