Speaking very generally, investment in capability-class supercomputers by national governments tends to scale with gross domestic product. …
HPE And AMD Bag The Big Supercomputer Deal Down Under was written by Timothy Prickett Morgan at The Next Platform.
Here we go again: Seven countries, including the U.S., U.K., Japan, and India, are again pushing tech companies to provide encryption backdoors for law enforcement, The Verge reports. The new international statement says encryption poses “significant challenges to public safety.” The U.S. and allies have long pushed for backdoors, even as security advocates have warned that criminals will find ways to exploit holes in encryption.
ISPs protest: ISPs in Dhaka, Bangladesh, are threatening to shut down service for three hours a day over a dispute on overhead wire replacement, Dhaka Tribune reports. The city is planning to move the wires underground, but ISPs are concerned about potential problems during the switch. Threatening a blackout to protest potential blackouts seems counterproductive.
Decency police: The government of Pakistan has banned short video app TikTok over “immoral and indecent” content, Al Jazeera says. It’s unclear what TikTok content the Pakistan Telecommunication Authority found offensive. A kickboxer in the country, hoping to market his training services on TikTok, challenged the ban, however, Reuters reports. “If TikTok can be banned even though it has millions of videos because of a few offensive ones, why can’t the whole internet be shut Continue reading
Today's Network Break examines a new security service from Cloudflare, a SonicWall vulnerability that affects a massive number of its appliances, a fresh funding round for networking startup Alkira, and more nerdy networking news.
The post Network Break 306: Cloudflare Launches Secure NaaS Offering; SonicWall Vulnerability Affects Nearly 800,000 Devices appeared first on Packet Pushers.
A few weeks ago we published an interesting discussion on network operating system details based on an excellent set of questions by James Miles.
Unfortunately we got so far into the weeds at that time that we answered only half of James’ questions. In the second Q&A session Dinesh Dutt and myself addressed the rest of them including:
And of course we couldn’t avoid the famous last question: “Should network engineers program network devices?”
You’ll need Standard or Expert ipSpace.net subscription to watch the videos.
A few weeks ago we published an interesting discussion on network operating system details based on an excellent set of questions by James Miles.
Unfortunately we got so far into the weeds at that time that we answered only half of James' questions. In the second Q&A session Dinesh Dutt and myself addressed the rest of them including:
And of course we couldn’t avoid the famous last question: “Should network engineers program network devices?”
You’ll need Standard or Expert ipSpace.net subscription to watch the videos.
The case for a learned sorting algorithm, Kristo, Vaidya, et al., SIGMOD’20
We’ve watched machine learning thoroughly pervade the web giants, make serious headway in large consumer companies, and begin its push into the traditional enterprise. ML, then, is rapidly becoming an integral part of how we build applications of all shapes and sizes. But what about systems software? It’s earlier days there, but ‘The case for learned index structures’(Part 1 Part 2), SageDB and others are showing the way.
Today’s paper choice builds on the work done in SageDB, and focuses on a classic computer science problem: sorting. On a 1 billion item dataset, Learned Sort outperforms the next best competitor, RadixSort, by a factor of 1.49x. What really blew me away, is that this result includes the time taken to train the model used!
Suppose you had a model that given a data item from a list, could predict its position in a sorted version of that list. 0.239806? That’s going to be at position 287! If the model had 100% accuracy, it would give us a completed sort just by running over the dataset and putting Continue reading
My friend Marjan Bradeško wrote a great article describing how we tend to forget common sense and rely too much on technology. I would strongly recommend you read it and start thinking about the choices you make when building a network with magic software-intent-defined-intelligent technology from your preferred vendor.
My friend Marjan Bradeško wrote a great article describing how we tend to forget common sense and rely too much on technology. I would strongly recommend you read it and start thinking about the choices you make when building a network with magic software-intent-defined-intelligent technology from your preferred vendor.
Today, we’re very excited to announce our plans for Cloudflare Intrusion Detection System, a new product that monitors your network and alerts when an attack is suspected. With deep integration into Cloudflare One, Cloudflare Intrusion Detection System gives you a bird’s eye view of your entire global network and inspects all traffic for bad behavior, regardless of whether it came from outside or inside your network.
Enterprises build firewall rules to keep their networks safe from external and internal threats. When bad actors try to attack a network, those firewalls check if the attack matches a rule pattern. If it does, the firewall steps in and blocks the attack.
Teams used to configure those rules across physical firewall appliances, frequently of different makes and models, deployed to physical locations. Yesterday, we announced Magic Firewall, Cloudflare’s network-level firewall delivered in our data centers around the world. Your team can write a firewall rule once, deploy it to Cloudflare, and our global network will protect your offices and data centers without the need for on-premises hardware.
This is great if you know where attacks are coming from. If you don’t have that level Continue reading
In early 2018 I described how Hans Verkerk implemented zero-touch provisioning with Ansible. Recently he rewrote his scripts as a Python-only solution using Nornir. Enjoy!
In early 2018 I described how Hans Verkerk implemented zero-touch provisioning with Ansible. Recently he rewrote his scripts as a Python-only solution using Nornir. Enjoy!
The NYPost Hunter Biden story has triggered a lot of sleuths obsessing on technical details trying to prove it's a hoax. So far, these claims are wrong. The story is certainly bad journalism aiming to misinform readers, but it has not yet been shown to be a hoax.
In this post, we look at claim the timelines don't match up with the manufacturing dates of the drives. Sleuths claim to prove the drives were manufactured after the events in question, based on serial numbers.
What this post will show is that the theory is wrong. Manufacturers pad warrantee periods. Thus, you can't assume a date of manufacture based upon the end of a warrantee period.
The story starts with Hunter Biden (or associates) dropping off a laptop at a repair shop because of water damage. The repair shop made a copy of the laptop's hard drive, stored on an external drive. Later, the FBI swooped in and confiscated both the laptop and that external drive.
The serial numbers of both devices are listed in the subpoena published by the NYPost:
Front Row with Frontera; Spectrum Scale Strategy; Computational Storage Deep Dive; Neural Programming Frontiers; MLPerf for HPC + More…
On today’s episode for the week, we talk to Dan Stanzione, director of the Texas Advanced Computing Center, to get an update on the “Frontera” supercomputer. …
Next Platform TV: October 16, 2020 was written by Nicole Hemsoth at The Next Platform.
The NYPost has an article on Hunter Biden emails. Critics claim that these don't look like emails, and that there are errors with the fonts, thus showing they are forgeries. This is false. This is how Apple's "Mail" app prints emails to a PDF file. The font errors are due to viewing PDF files within a web browser -- you don't see them in a PDF app.
In this blogpost, I prove this.
I'm going to do this by creating forged email. The point isn't to prove the email wasn't forged, it could easily have been -- the NYPost didn't do due diligence to prove they weren't forged. The point is simply that that these inexplicable problems aren't evidence of forgery. All emails printed by the Mail app to a PDF, then displayed with Scribd, will look the same way.
To start with, we are going to create a simple text file on the computer called "erratarob-conspire.eml". That's what email messages are at the core -- text files. I use Apple's "TextEdit" app on my MacBook to create the file.
