VMware has released a new version of its vSphere-friendly configuration of OpenStack, the popular open source cloud platform.
VMware Integrated OpenStack (VIO) 2.5 hit general availability yesterday. VMware calls it “…the easiest and fastest route to build an OpenStack cloud on top of vSphere, NSX and Virtual SAN.”
According to a blog posting by Pete Cruz, the chief upgrades in the new version include easier integration with the underlying vSphere stack; a less cumbersome management control plane, …
The Elastic stack is a versatile collection of open source software tools that make gathering insights from data easier. Formerly referred to as the ELK stack (in reference to Elasticsearch, Logstash, and Kibana), the growing list of tools that integrate with the platform (such as Beats) have outgrown the acronym but provide ever-growing capability for users and developers alike.
At the upcoming Southeast Linuxfest 2016, I’ll be covering some of the steps to get started using each of these parts of the stack. In this article, we’ll look at each in turn to summarize the capabilities, requirements, and interesting use cases that apply to each.
The idea for the Qt Automotive Suite was born when The Qt Company, Pelagicore and KDAB sat down and shared their experiences of projects using Qt for In-vehicle Infotainment (IVI). With cumulative experience from over 20 automotive projects it was noted how Qt is really well suited to the needs of building IVIs and Instrument Clusters, that there were already millions of vehicles on the road with Qt inside, and that there were a lot of ongoing projects. There was though a feeling that things could be even better, that there were still a few things holding back the industry, contributing to the sense that shipped IVI systems could be built faster, cheaper and with a higher quality.
One observation was that additional infrastructure components and tooling were being created. While it is great to see software being built on top of Qt, from an industry perspective it is inefficient with duplication of work, little reuse across projects and engineering resources being used to maintain them rather than focusing on differentiating features. So we’ve added some of these components to the Qt Automotive Suite and will continue to add more over time.
To overcome the limitations of and to increase the security mechanisms provided by standard ugo/rwx permissions and access control lists, the United States National Security Agency (NSA) devised a flexible Mandatory Access Control (MAC)…
[[ This is a content summary only. Visit my website for full links, other content, and more! ]]
Apache, Apache Mesos, and Mesos are either registered trademarks or trademarks of the Apache Software Foundation (ASF) in the United States and/or other countries. MesosCon is run in partnership with the ASF.
Setting up a motion detection system on Linux is fairly easy and simple. All that we need is a webcam (or laptop), the motion package, and a few minutes to set everything up. The purpose for doing this may be private space surveillance, enhancement of personal security, or simply a fun project. Whatever the case, this quick guide is not intended to promote illegal activities such as unauthorized video recording of people and their activities. That said, please use the knowledge offered here with ethical conduct.
Kris Buytaert is known as one of the instigators of the current DevOps movement and organizer of several related conferences, including DevOpsDays and Config Management Camp. He is a long-time Linux and open source consultant who often claims that everything is a freaking DNS problem. You can find him speaking at events and consulting as the CTO (Chief Trolling Officer) at Inuits on everything from Infrastructure as Code to Continuous Delivery.
Kris Buytaert is an organizer of DevOpsDays and Config Management Camp.
Linux.com: Why are so many organizations embracing DevOps?
Kris Buytaert: A lot of organizations want to be the cool kids on the block. They want to be like the DevOps poster children: the organizations that don’t have problems deploying new functionality for their users, that don’t have problems with stability, and that are looked at as nice places to work.
DevOps promises higher throughput for software delivery, a more stable platform, and better security. Doesn’t everyone want that?
But it was a trick question wasn’t it? Isn’t the real question, “why do so many organizations claim they are embracing DevOps, but none of them really are embracing it”?
Because DevOps is hard for most organizations, it requires a drastic change in culture, management style, organization, and software architecture. Usually management isn’t ready to make those changes, so they create a new silo in their organization called the DevOps team, put in place a semi automated deployment pipeline with no tests, implement some form of automation but keep doing manual changes in production, and implement standups so they can call their development methods agile too.
Linux.com: Why are individuals interested in participating?
Kris: A lot of individuals see the above mentioned goals as something that would reduce the workload, risk and stress that comes with their job. They see automation through collaboration as an option to reduce the risk, reduce the downtime, and get a better night’s sleep with the side effect of taking away the dull parts of their work, the manual repetitive work, to give them more time to focus on the aspects of their work they never had time for: e.g. really looking at how the application behaves, how users (ab)use it, or improving performance and security. So, they stand up and start the fight to improve their environment.
Linux.com: How can you tell if a company has truly adopted DevOps?
Kris: What does it mean to truly adopt DevOps? Is it 5 deploys a day? 1000 deploys? Is it having security embedded in your pipeline? It’s none of the above. DevOps is a journey. I don’t think you are ever done adopting DevOps. DevOps, much like security, is a lifestyle. If there is one thing that won’t change in today’s technology industry, it is the fact that change will happen all the time, so you’d better be prepared to adapt. An organisation that has figured out that it will be learning and will need to continuously be improving collaboration between people with different skills in order to improve their quality, whether they call it DevOps or not, is probably one that has truly adopted DevOps.
Linux.com: What is the overwhelming hurdle?
Kris: People and inertia. A lot of organizations have trained people for years to do X, and when those people suggested they do Y, or tried to do Z, many of them got “no” as an answer, or even worse, they got punished for it. We have a whole generation of people that are trained like this and now we want to see the opposite behaviour. Corporations have taught engineers not to solve problems for themselves, but to buy a 3rd party as insurance. However, this also won’t solve their problem; it’s just becomes not their problem anymore (often called proprietary software). And now, we want self-supporting teams that take responsibility for what they build and ship and who are expected to run and manage it themselves. It’s going to take a while for those people to break their habits and for their management to actually encourage them to change when the managers themselves will also need to change. That is the biggest hurdle for most organizations. It’s going to take a long time for some organizations to change, if they change at all.
Linux.com: What advice would you give to people who want to get started in DevOps?
Kris: The first part of my advice would be to find a local DevOps meetup or a local DevOpsDays event. These are put on by the community to teach people about DevOps. You’ll meet people there with experience and with ideas, but you’ll also meet other newcomers.
Secondly, a story told way too often is that the DevOps movement started out of the open source world back at the very first DevOpsDays in Ghent in 2009. Most of the people present had a strong background in open source, and most of the success stories were from people leveraging the power of these open source tools, but also the mindset of collaboration came out of the open source community.
So the best advice I can give to people is to start looking into those open source projects, yes there will be vendors trying to catch your attention, uh money, by claiming their 15 year-old legacy tool now is DevOps ready, but really everything you need is open.
Learn more about DevOps from leaders in the field.
Read this Q&A with Michael Ducy, co-host of the Goat Farm podcast and blog and Chef’s Manager of Solutions Architects.
Read this Q&A with Patrick Debois, best known as the founder of DevOpsDays and as a creator of the DevOps movement, which explains why some refer to him as the “Godfather of DevOps.”
IBM recently announced the Data Science Experience on its Bluemix cloud platform to help developers build intelligent applications and make data analytics more accessible to the enterprise.
One of the main themes of this year’s Embedded Linux Conference and OpenIoT Summit was the challenge of bridging the growing number of Internet of Things (IoT) standards. Many speakers were hopeful about the potential for achieving functional interoperability, if not a unifying standard, and there were even calls for a possible merger between two of the largest open source efforts: AllSeen and IoTivity.
Others, however, like Dutch IT consultant Benedikt Herudek, were more skeptical. “Forget about standards,” Herudek told attendees to his presentation “IoT Consensus — A Solution Suggestion to the ‘Baskets of Remote’ Problem.”
“Sometimes it works, but in my experience getting everybody to use the same standard is usually wishful thinking,” he said.
Herudek is not content with an IoT world separated into silos, however. He has a potential solution based on a proven technology: Bitcoin. In Herudek’s view, the same qualities of universality and security that mark Bitcoin and its blockchain technology for financial transactions could be applied to the IoT messaging protocol problem. Specifically, Herudek outlines a version of the Bitcoin’s “proof of work” consensus mechanism that would translate messages between any application without requiring a trusted middleman or resorting to industry or vendor protocol standards.
The scheme is inspired in part from IBM’s ADEPT proof-of-concept draft, developed with Samsung, which similarly uses Bitcoin and blockchain for IoT. According to Herudek, his proposal is an improvement over ADEPT. Both solutions are based on the peer-to-peer Bitcoin cryptocurrency protocol for financial transactions, which offers the equivalents of IoT endpoints with its “wallets” and “miners.” The blockchain, meanwhile, acts as a public distributed ledger.
“Bitcoin and blockchain can create consensus between a lot of different parties on the state of transactions,” said Herudek. “Blockchain has some economic advantages over a centralized cloud, which maybe after the next big upgrade won’t support your device anymore. A distributed blockchain system, which is kind of self-organizing and not dependent on one vendor, could give you the advantage of having something long lasting. Bitcoin lets you achieve security in a distributed system without identity and access management. Typically, if people misbehave, the identity management system will kick you out, but Bitcoin doesn’t do that because it says: ‘I have such a convincing algorithm, nobody can cheat.’”
Instead of using Bitcoin’s “proof of work,” in which miners compete to be the first to place transactions into blocks, Herudek proposes using a “proof of useful work.” He uses an IoT example of a smartphone telling a smart fridge to open the door.
“This occurs at the very highest layer, not the transportation layer,” says Herudek. “You have to get an agreement on the message format and the semantics — which value can you put into which slot — and you need to achieve a certain action. So, instead we can take a model similar to Bitcoin in which you have a translator, something like a miner, which is responsible for delivering message formats. You would have a three-step process: a format handshake, content handshake, and an action handshake.”
Replacing Hashing with Smart Learning Algorithms
The challenges of using Bitcoin for IoT include using miner hashing sessions for very well-defined use cases, like transferring money between two parties. “I don’t think that is something you can transfer to the number of use cases in IoT,” said Herudek.
“Instead of using hashing, you could instead use something like a machine-learning algorithm to translate, and then give rewards for translating,” continued Herudek. “The algorithm would essentially use blockchain as a training set. You would not only record the transactions between entities, but also the whole history of how they tried to do the handshakes. You could give them features like which vendor is participating and which technology is used, or apply inputs like noticing if someone is standing in front of the fridge. You can facilitate this by adding a training set to the blockchain.”
From there, Herudek discussed issues such as immutability, security, and how to measure “effort.” Much of the proposal requires a solid understanding of both Bitcoin and the latest IoT message interchange methods. The idea is intriguing, however, as it might emerge as a ready-made alternative to a standards development process that could take many years to achieve on both the technological and political/business levels.
“In the world of IT, it can take years to integrate apps,” said Herudek. “Yet, if you open your Bitcoin wallet on your iPhone, you can communicate within minutes to any other wallet in the world.”
More information is available in Herudek’s Service Technology Magazine article on using Bitcoin and Blockchain for IoT messaging, as well as this slide set from his presentation.
