TensorFlow.js: machine learning for the web and beyond

TensorFlow.js: machine learning for the web and beyond Smilkov et al., SysML’19

If machine learning and ML models are to pervade all of our applications and systems, then they’d better go to where the applications are rather than the other way round. Increasingly, that means JavaScript – both in the browser and on the server.

TensorFlow.js brings TensorFlow and Keras to the the JavaScript ecosystem, supporting both Node.js and browser-based applications. As well as programmer accessibility and ease of integration, running on-device means that in many cases user data never has to leave the device.

On-device computation has a number of benefits, including data privacy, accessibility, and low-latency interactive applications.

TensorFlow.js isn’t just for model serving, you can run training with it as well. Since it’s launch in March 2018, people have done lots of creative things with it. And since it runs in the browser, these are all accessible to you with just one click! Some examples:

Juniper Syslog

3 steps to configure Syslog. Define a logging policy Define remote logging servers Define a logging source address (optional) Configuration Log to a local file. Logs are stored in the /var/log directory. Define a logging policy. cmd set system syslog user * any...continue reading

Emulating Juniper Devices – Various options

Hi,

I have got a lot of requests for writing up a blog post on various Methods of emulating Juniper devices for practice.

Note : For 2/3 methods to work, you need to have official Junos software (vmx-vcp and vmx-vfp)

Method 1 – Gns3

Most popular and Familiar Method – Install via gns3

After installing Gns3, download the

Vmx-vfp Appliance – https://docs.gns3.com/appliances/juniper-vmx-vfp.html

Vmx-vcp Appliance – https://docs.gns3.com/appliances/juniper-vmx-vcp.html

Documentation is straight forward, all you need to do is double click on the appliance and upload the image when requested, as easy as it can get.

 

Method2 – Vagrant

Use Juniper uploaded Images via vagrant.

Most of Juniper Vqfx / JNCIE-DC practice came up from a vagrant, You don’t need to have any official images or access to Juniper software downloads. The downside is that it only supports VQFX and generic SRX, but for most of the Routing protocol and MPLS learning this should be good.

https://app.vagrantup.com/boxes/search?utf8=%E2%9C%93&sort=downloads&provider=&q=juniper

https://app.vagrantup.com/juniper/boxes/vqfx10k-re

https://app.vagrantup.com/juniper/boxes/vqfx10k-pfe

Again, following Github link, will auto setup the topology without you worrying about much details to vagrant.

https://github.com/Juniper/vqfx10k-vagrant -> Go into a specific folder and say vagrant up, that should take Continue reading

Helping To Build Cloudflare, Part 2: The Most Difficult Two Weeks

This is part 2 of a six part series based on a talk I gave in Trento, Italy. Part 1 is here.

It’s always best to speak plainly and honestly about the situation you are in. Or as Matthew Prince likes to put it “Panic Early”. Long ago I started a company in Silicon Valley which had the most beautiful code. We could have taught a computer science course from the code base. But we had hardly any customers and we failed to “Panic Early” and not face up to the fact that our market was too small.

Ironically, the CEO of that company used to tell people “Get bad news out fast”. This is a good maxim to live by, if you have bad news then deliver it quickly and clearly. If you don’t the bad news won’t go away, and the situation will likely get worse.

Cloudbleed

Cloudflare had a very, very serious security problem back in 2017. This problem became known as Cloudbleed. We had, without knowing it, been leaking memory from inside our machines into responses returned to web browsers. And because our machines are shared across millions of web sites, that meant that HTTP requests Continue reading

Build a network emulator using Libvirt

I want to explore some of the network virtualization and emulation building blocks available on a Linux system. In this post, I create a simple network emulation scenario using Libvirt, the Qemu/KVM hypervisor, and Linux bridges to create and manage interconnected virtual machines on a host system.

Libvirt provides a command-line interface that hides the low-level virtualization and networking details, enabling one to easily create and manage virtual networking scenarios. It is already used as a basis for some existing network emulators, and other applications and tools. It is available in almost every Linux distribution.

The network emulation scenario

As you work through the examples in this post, you will create a very simple network topology which is intended to demonstrate the use of Libvirt and other virtualization tools to build a network emulator and is not intended to emulate a real-world network. However, once you understand its operation, you may use Libvirt to create large, complex network topologies intended to emulate real-world network scenarios.

The example I created for this post consists of three virtual machines serving as routers connected to each other in a ring topology. On each side of this emulated network, you will create Continue reading

When Does AWS Break Through $100 Billion?

While there is no question that Amazon Web Services is the largest and most profitable public cloud on planet earth, and that it provides the best subsidy imaginable for a cut-throat online retail business that is its parent company, AWS has not taken over the entire IT world any more than Amazon has become the sole place to buy stuff. …

When Does AWS Break Through $100 Billion? was written by Timothy Prickett Morgan at .

Scraping Envoy Metrics Using the Prometheus Operator

On a recent customer project, I recommended the use of Heptio Contour for ingress on their Kubernetes cluster. For this particular customer, Contour’s support of the IngressRoute CRD and the ability to delegate paths via IngressRoutes made a lot of sense. Of course, the customer wanted to be able to scrape metrics using Prometheus, which meant I not only needed to scrape metrics from Contour but also from Envoy (which provides the data plane for Contour). In this post, I’ll show you how to scrape metrics from Envoy using the Prometheus Operator.

First, I’ll assume that you’ve already installed and configured Prometheus using the Prometheus Operator, a task which is already fairly well-documented and well-understood. If this is something you think would be helpful for me to write a blog post on, please contact me on Twitter and let me know.

The overall process looks something like this:

  1. Modify the Envoy DaemonSet (or Deployment, depending on your preference) to add a sidecar container and expose additional ports.
  2. Modify the Service for the Envoy DaemonSet/Deployment to expose the ports you added in step 1.
  3. Add a ServiceMonitor object (a CRD added by the Prometheus Operator) to tell Prometheus to scrape Continue reading

Weekend Reads 020119

I think that the data center industry has reached a point of equilibrium and stabilization where the players for cloud service have been defined, the large mega applications (like Facebook and Google) have become a monopoly on their respective businesses, and there has been slow but sustained growth in all of these markets. —Yuval Bacher

How many times in the software industry have you heard developers fight over which programming language is better? How many times have you heard about that one pattern that will fix all of our problems moving forward? —Marco Palladino

What’s clear is that supply chain attacks are a well-established, if underappreciated, method of surveillance — and much work remains to be done to secure computing devices from this type of compromise. —Micah Lee, Henrik Moltke

But when designing underlying cyber protections, too many architects are taking zero trust to be the primary objective. This is a misinterpretation. In the first instance, we should look to protect our resources from attack. What zero trust reminds us is that we are fallible, and that we should put in place backup plans in the form of monitoring and incident response for the (hopefully rare) cases where our protection Continue reading

Advanced Solutions Customer Story Part 1: Why NSX-T?

 

Customer Overview

Advanced Solutions, a DXC Technology company, was formed in 2004 and employs about 500 staff to support the government of the Canadian province of British Columbia and other public sector customers with IT and business process solutions. For government agencies and services to continue operating efficiently and effectively, it is essential that the IT resources that they require are provided quickly and accurately.

Key Pain Points

All IT organizations are acutely familiar with the wide range of pain points and obstacles that can stand in the way of delivering resources to empower their businesses to move with speed and agility. One of the most common hindrances to IT, and therefore business agility is painfully slow provisioning processes, which can take weeks just to provision an application. The most common bottleneck within these processes is provisioning networking and security services. This is a key pain point for Advanced Solutions, but one that VMware is helping them solve with the VMware NSX Data Center network virtualization platform.

Dan Deane, Solutions Lead at Advanced Solutions says, “The key IT pain points that VMware solutions are helping us solve are around networking and provisioning.”

New Use Cases

Advanced Solutions was Continue reading

Check Point GAiA as Personal Firewall on Linux – Part1

Check Point GAiA is the next generation Secure Operating System for all Check Point appliances, open servers and virtualized gateways. In this tutorial we will create a network infrastructure which supports usage of Gaia Qemu VM as a personal firewall on Ubuntu Linux. We will also go through the entire installation of Gaia on Qemu VM. This firewall appliance can be used up to 15 days period covered by a free trial Gaia license (no registration needed).

Hardware:  Asus K55VM laptop:
- Intel(R) Core(TM) i7-3610QM CPU @ 2.30GHz
- RAM - 2 x Kingston DDR3 8192MB,
- HDD - ST1000LM024 HN-M101MBB 1000GB

Hardware requirements:
- Gaia QEMU VM: RAM - 4096MB, HDD - 15GB
- Windows 7 QEMU VM: RAM 2048MB, HDD - 15 GB

Software:
Host - Kubuntu Linux 18.04.1 LTS with installed QEMU emulator version 3.0.0 and KVM module
Guest 1 - Checkpoint GAiA R80.10, OS build 462, OS kernel version 2.6.18-92cpx86_64
Guest 2 - Windows 7 Home Premium, x86 with installed Smart Console R80.10 Build 991140073

Credentials - username/password:
- Gaia web portal: admin/check123point
- Gaia expert mode: check123point
- Windows 7: no password Continue reading

Helping To Build Cloudflare, Part 3: Audacity, Diversity and Change

This is part 3 of a six part series based on a talk I gave in Trento, Italy. To start from the beginning go here.

After Cloudbleed, lots of things changed. We started to move away from memory-unsafe languages like C and C++ (there’s a lot more Go and Rust now). And every SIGABRT or crash on any machine results in an email to me and a message to the team responsible. And I don’t let the team leave those problems to fester.

Making 1.1.1.1

So Cloudbleed was a terrible time. Let’s talk about a great time. The launch of our public DNS resolver 1.1.1.1. That launch is a story of an important Cloudflare quality: audacity. Google had launched 8.8.8.8 years ago and had taken the market for a public DNS resolver by storm. Their address is easy to remember, their service is very fast.‌‌

But we thought we could do better. We thought we could be faster, and we thought we could be more memorable. Matthew asked us to get the address 1.1.1.1 and launch a secure, privacy-preserving, public DNS resolver in a couple of months. Continue reading

1 … 1,358 1,359 1,360 1,361 1,362 … 3,872