Minimizing trial and error in the drug discovery process – Microsoft Research

molecules, stock image

In 1928, Alexander Fleming accidentally let his petri dishes go moldy, a mistake that would lead to the breakthrough discovery of penicillin and save the lives of countless people. From these haphazard beginnings, the pharmaceutical industry has grown into one of the most technically advanced and valuable sectors, driven by incredible progress in chemistry and molecular biology. Nevertheless, a great deal of trial and error still exists in the drug discovery process. With an estimated space of 1060 small organic molecules that could be tried and tested, it is no surprise that finding useful compounds is difficult and that the process is full of costly dead ends and surprises.

The challenge of molecule design also lies at the heart of many applications outside pharmacology, including in the optimization of energy production, electronic displays, and plastics. Each of these fields has developed computational methods to search through molecular space and pinpoint useful leads that are followed up in the lab or in more detailed physical simulations. As a result, there are now vast libraries of molecules tagged with useful properties. The abundance of data has encouraged researchers to turn to data-driven approaches to reduce the degree of trial and error in chemical development, and the aim of our paper being presented at the 2018 Conference on Neural Information Processing Systems (NeurIPS) is to investigate how recent advances, specifically in deep learning techniques, could help harness these libraries for new molecular design tasks.

Deep learning with molecular data

Figure 1: The chemical structure of naturally occurring penicillin (penicillin G) and its representation as a graph in a GGNN. The messages passed in the environment of a single node are shown as curved arrows, and the neural networks that transform the messages are shown as small squares. Repeated rounds of message passing allow each node to learn about its surroundings (gray circles).

Deep learning methods have revolutionized a range of applications requiring understanding or generation of unstructured data such as pictures, audio, and text from large datasets. Applying similar methods to organic molecules poses an interesting challenge because molecules contain a lot of structure that is not easy to concisely capture with flat text strings or images (although some schemes do exist). Instead, organic chemists typically represent molecules as a graph where nodes represent atoms and edges represent covalent bonds between atoms. Recently, a class of methods that have collectively become known as neural message passing has been developed precisely to handle the task of deep learning on graph-structured data. The idea of these methods is to encode the local information, such as which element of the periodic table a node represents, into a low-dimensional vector at each node and then pass these vectors along the edges of the graph to inform each node about its neighbors (see Figure 1). Each message is channeled through small neural networks that are trained to extract and combine information to update the destination node’s vector representation to be informative for the downstream task. The message passing can be iterated to allow each node to learn about its more distant neighbors in the graph. Microsoft Research developed one of the earliest variants of this class of deep learning models—the gated graph neural network (GGNN). Microsoft’s primary application focus for GGNNs is in the Deep Program Understanding project, where they are used to analyze program source code (which can also be represented using graphs). Exactly the same underlying techniques are applicable to molecular graphs.

Generating molecules

Figure 2: Example molecules generated by our system after being trained on organic solar cell molecules (CEP database).

Broadly speaking, there are two types of questions that a machine learning system could try to solve in molecule design tasks. First, there are discriminative questions of the following form: What is the property Y of molecule X? A system trained to answer such questions can be used to compare given molecules by predicting their properties from their graph structure. Second, there are generative questions—what is the structure of molecule X that has the optimum property Y?—that aim to invent structures that are similar to molecules seen during training but that optimize for some property. The new paper concentrates on the latter, generative question; GGNNs have already shown great promise in the discriminative setting (for example, see the code available here).

The basic idea of the generative model is to start with an unconnected set of atoms and some latent “specification” vector for the desired molecule and gradually build up molecules by asking a GGNN to inspect the partial graph at each construction step and decide where to add new bonds to grow a molecule satisfying the specification. The two key challenges in this process are ensuring the output of chemically stable molecules and designing useful low-dimensional specification vectors that can be decoded into molecules by the generative GGNN and are amenable to continuous optimization techniques for finding locally optimal molecules.

For the first challenge, there are many chemical rules that dictate whether a molecular structure is stable. The simplest are the valence rules, which dictate how many bonds an element can make in a molecule. For example, carbon atoms have a valency of four and oxygen a valency of two. Inferring these known rigid rules from data and learning to never violate them in the generative process is a waste of the neural network’s capacity. Instead, in the new work, we simply incorporate known rules into the model, leaving the network free to discover the softer trends and patterns in the data. This approach allows injection of domain expertise and is particularly important in applications where there is not enough data to spend on relearning existing knowledge. We believe that combining this domain knowledge and machine learning will produce the best methods in the future.

Figure 3: Example molecule optimization trajectory when optimizing the quantitative estimate of drug-likeness (QED) of a molecule after training on the ZINC database. The initial molecule has a QED of 0.4, and the final molecule has a QED of 0.9

Figure 3: Example molecule optimization trajectory when optimizing the quantitative estimate of drug-likeness (QED) of a molecule after training on the ZINC database. The initial molecule has a QED of 0.4, and the final molecule has a QED of 0.9

For the second challenge, we used an architecture known as a variational autoencoder to discover a space of meaningful specification vectors. In this architecture, a discriminative GGNN is used to predict some property Y of a molecule X, and the internal vector representations in this discriminative GGNN are used as the specification vector for a generative GGNN. Since these internal representations contain information about both the structure of molecule X and the property Y, continuous optimization methods can be used to find the representation that optimizes property Y; the representation is then decoded to find useful molecules. Example molecules generated by the new system are shown in Figures 2 and 3.

Collaborating with experts

The results in the paper are very promising on simple molecule design tasks. However, deep learning methods for molecule generation are still in their infancy, and real-world molecule design is a very complicated process with many different objectives to consider, such as molecule efficacy, specificity, side effects, and production costs. To make significant further progress will require collaboration of machine learning experts and expert chemists. One of the main aims of this paper is to showcase the basic capabilities of deep learning in this space and thereby act as a starting point for dialogue with chemistry experts to see how these methods could enhance their productivity and have the most impact.

Go to Original Article
Author: Steve Clarke

How to ask questions the way great business leaders do: CFO Summit

NEWTON, Mass. — To succeed in an environment roiled by new digital business models, asking questions matters more than having all the answers. But not just any questions. The most successful business leaders ask questions that are designed to prove them wrong, are likely to make them uncomfortable, and will drive them to reflect on their behaviors and beliefs. That’s how they become great problem-solvers.

That was the motivational message to CFOs from Hal Gregersen, executive director of the MIT Leadership Center, at the recent MIT Sloan CFO Summit. Certainly, senior financial executives understand the importance of asking questions in their line of work, Gregersen conceded to the CFO Summit audience. But, “given this trend of digital transformation, which is changing all of our lives,” CFOs need to ask better questions — and more of them — if they aim to play a strategic role in this period of business upheaval, he said.

Drawing on research from his new book, Questions Are the Answer: A Breakthrough Approach to Your Most Vexing Problems at Work and in Life, Gregersen served up a who’s who of  “certified, focused problem-solvers” to make his point — from the founder of JetBlue to the CEO of Patagonia, to Fadi Ghandour, co-founder of Aramex, a logistics company headquartered in Dubai and the first United Arab Emirates-based company to be listed on an American Stock Exchange.

“Hired by FedEx to run the business in [Arab hotspots], he got so good at it that he started a competitor,” Gregersen said of Ghandour, who stepped down from Aramex in 2012 and now oversees a technology venture capital fund.

Passive data

According to Gregersen, a big factor in Aramex’s success was Ghandour’s ability to ask questions that elicited “passive data,” or data that he wouldn’t ordinarily hear. When Ghandour traveled for business — sometimes arriving late — instead of making a chauffeured beeline to the hotel for some rest, he insisted on being picked up by one of his Aramex truck drivers.

“They know him, he knows them, [and] they speak candidly and pass along data that isn’t coming through official channels.”

MIT CFO Summit, great business leaders, digital transformation, CFO
Hal Gregersen, director of the MIT Leadership Center, gives a talk on how to ask better questions at the MIT Sloan CFO Summit.

Getting out of the corporate bubble where people are likely to tell you what you want to hear is crucial to good decision-making, Gregersen said. “It’s the passive data that is not getting pushed at us that allows us to figure out what we know we don’t know,” he told the CFO Summit audience.

According to Gregersen, only about 60% of the decisions made by great business leaders are right, but what sets them apart from their less high-achieving peers is they catch the wrong decisions faster.

“And the only way to catch those wrong decisions faster is by actively seeking passive data,” he said.

Soul-searching reflection

Driving around also figured large in another business leader profiled by Gregersen during his CFO Summit talk.

MIT CFO Summit, great business leaders, digital transformation, CFO
At the MIT Sloan CFO Summit, Hal Gregersen’s slide on three conditions that help business leaders ask better questions.

Rose Marcario, now the CEO of Patagonia, was a private equity dealmaker en route to a big meeting in New York when her driver slammed on the brakes. Annoyed that she was no longer making progress, Marcario looked out the window and saw the culprit was a hobbled pedestrian slowly crossing the street. The sight “struck her like a dagger in the heart,” Gregersen said. The woman resembled her mother, who had suffered from psychological problems when Marcario was growing up.

Her reaction to this encounter is an example of another condition of asking great questions, according to Gregersen: being reflective. Marcario asked the driver to drop her off in Central Park, and as she walked, Gregersen said she clarified the problem she would solve five years later when she joined Patagonia as its CFO: “‘How can I make a living without losing my soul?'”

How Pixar creates great business leaders

The third condition observed by great business leaders is they put themselves in situations where they are susceptible to being told they are wrong, Gregersen said. At Pixar, directors participate in the Braintrust meetings, where they meet to discuss the current movie Pixar is working on.

The meetings are an invention of Disney and Pixar Animation Studios President Ed Catmull, a computer scientist by training, and they can be “devasting, difficult, but absolutely necessary, ” Gregersen told the CFO Summit audience. 

[We need] to regularly put ourselves in situations where we are wrong and feel uncomfortable. When we do that, that is when the questions come that literally unlock answers we ordinarily wouldn’t get.
Hal Gregersenexecutive director of the MIT Leadership Center

In his research at Pixar, Gregersen learned that some of the best movies made there are based on the life stories of the directors. So, the Braintrust sessions are not only a critique of the movie’s merits, but also a judgment of sorts on the director’s life story.

“Top-of-the-line leaders are getting feedback [in a setting] where they are wrong, uncomfortable, and they are forced to shut up and listen,” Gregersen said. 

The point of the anecdote Gregersen told the CFO Summit attendees was not that they go back to their offices and set up Braintrust meetings, but that they find ways to put themselves and their teams in situations where they can get feedback of that brutal caliber.

“As we walk into that world of digital transformation that’s full of opportunity for us, the best advantage we can take of it is by … figuring out our own path to regularly put ourselves in situations where we are wrong and feel uncomfortable,” Gregersen said. “When we do that, that is when the questions come that literally unlock answers we ordinarily wouldn’t get.”

Go to Original Article
Author:

For Sale – Dell poweredge T410 plus i5 6600k pc 16GB DDR4

Discussion in ‘Desktop Computer Classifieds‘ started by moores211, Nov 18, 2018.

  1. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    Dell poweredge t410
    running windows server 2008
    4 x 1TB SAS drives
    1 x 450GB Cheeter 15K
    Xeon E5620 2.4GHz quad core
    4GB RAM
    £60

    next pc
    i5 6600k
    16GB hyperx savage DDR4
    ASUS Z170-P Motherboard
    after market CPU cooler
    Corsair VS650 PSU
    think of the case as a freebie as the sleds are missing for the hard drives and it’s a cheap thing anyway.
    £220

    Price and currency: 50 220
    Delivery: Delivery cost is not included
    Payment method: CASH BANK TRANSFER
    Location: bolton
    Advertised elsewhere?: Advertised elsewhere
    Prefer goods collected?: I prefer the goods to be collected

    ______________________________________________________
    This message is automatically inserted in all classifieds forum threads.
    By replying to this thread you agree to abide by the trading rules detailed here.
    Please be advised, all buyers and sellers should satisfy themselves that the other party is genuine by providing the following via private conversation to each other after negotiations are complete and prior to dispatching goods and making payment:

    • Landline telephone number. Make a call to check out the area code and number are correct, too
    • Name and address including postcode
    • Valid e-mail address

    DO NOT proceed with a deal until you are completely satisfied with all details being correct. It’s in your best interest to check out these details yourself.

    Attached Files:

  2. Eddie Twadds

    Eddie Twadds

    Active Member

    Joined:
    Nov 13, 2008
    Messages:
    94
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Brighouse
    Ratings:
    +7

    Couple of Qs about the I5 pc if that’s OK?

    Obviously no HDD but does everything work OK? Have never built a PC before so presume it’s just a case of adding HDD and Win10 OS from my existing build? Does the board come with software/instructions etc. Also, can you upload more pics of the case? Thanks.

  3. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    everything works great, you’re exactly right add hdd/sdd and you are away. no software/instructions but these can be downloaded from the asus website. pic uploaded

  4. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23
  5. Eddie Twadds

    Eddie Twadds

    Active Member

    Joined:
    Nov 13, 2008
    Messages:
    94
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Brighouse
    Ratings:
    +7

    Thanks for info. Buying on behalf of son number 2 so he’s budget limited. Any chance of £160 collected. Appreciate it’s cheeky but….

  6. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    sorry, I’d go to £200 failing that I’m splitting it
    i5 6600k £110
    ram £100
    motherboard £30
    psu £20
    case and cooler £10

  7. Eddie Twadds

    Eddie Twadds

    Active Member

    Joined:
    Nov 13, 2008
    Messages:
    94
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Brighouse
    Ratings:
    +7

    OK – would go to £200 if you can include snail mail delivery?

  8. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    sorry I’d prefer pick up and i really wouldn’t goo under 200.

  9. Eddie Twadds

    Eddie Twadds

    Active Member

    Joined:
    Nov 13, 2008
    Messages:
    94
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Brighouse
    Ratings:
    +7

    Understood. Can’t get to Bolton for a bit (based in Brighouse). Offer still on table if no other interest. GLWS.

  10. Dave26

    Active Member

    Joined:
    Dec 15, 2006
    Messages:
    297
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Manchester
    Ratings:
    +13

    Hi mate – cheeky offer of £190 and can sort out pick up over the weekend as I’m local?

  11. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    I’d go 190 without the psu(i could use it in another build)

  12. Dave26

    Active Member

    Joined:
    Dec 15, 2006
    Messages:
    297
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Manchester
    Ratings:
    +13

    I’ll stop being tight and agree £200

  13. moores211

    moores211

    Active Member

    Joined:
    Dec 28, 2008
    Messages:
    527
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    18
    Ratings:
    +23

    ha no probs – pm

Share This Page

Loading…

Go to Original Article
Author:

Windows 10 Tip: Name your tile folders – Windows Experience Blog

Did you know you can now name your tile folders, thanks to the Windows 10 October 2018 Update?
To create a tile folder in Start, just drag one tile on top of another for a second, then release. Continue dropping as many tiles into the folder as you’d like. When you expand the folder, you’ll see a new option to name it.
Check it out in action:

If you like this, check out more Windows 10 Tips.

Redesigning the Office App Icons to Embrace a New World of Work


Redesigning the Office App Icons to Embrace a New World of Work

Design is becoming the heart and soul of Office. Learn how we evolved our visual identity to reflect the simple, powerful, and intelligent experiences of Office 365.

Whoever said that nothing is more intimidating than the blank page probably never faced a redesign.

The last time we updated the Microsoft Office icons was in 2013, when selfies were new enough to become Oxford Dictionaries’ Word of the Year and emojis were new enough to be considered buzzworthy.

Clearly, a lot has changed since then — including how people get things done.

Over 1 billion people from vastly different industries, geographies, and generations use Office. They work on different platforms and devices and in environments that are faster, more distracting, and more connected than ever before.

To support this changing world of work, Office is transforming into a collaborative suite that lets you work together in real-time from almost any device. We’ve infused our tools with powerful AI: you can get insights from data with less effort, write a paper using your voice, or make your resume using LinkedIn insights. We’ve also added totally new apps to the suite like our AI-powered meetings and chat service, Microsoft Teams. In the end, it’s great design that makes these experiences fluid and seamless.

As a signal to our customers, we’ve evolved our Office icons to reflect these significant product changes. We’re thrilled to share the new icons for Office 365 with you today and tell the story behind their creation.

Carefully crafted designs that honor heritage and welcome the future

From the get-go, we embraced Office’s rich history and used it to inform design decisions. Strong colors have always been at the core of the Office brand, and new icons are a chance to evolve our palette. Color differentiates apps and creates personality, and for the new icons we chose hues that are bolder, lighter and friendlier — a nod to how Office has evolved.

We also used gestalt principles to further emphasize key product changes. Simplicity and harmony are key visual elements that reflect the seamless connectivity and intuitiveness of Office apps. While each icon has a unique and identifiable symbol, there are connections within each app’s symbol and the collective suite.

Flexible visual systems that work across platforms, devices, and generations

Today’s workforce includes five generations using Office on multiple platforms and devices and in environments spanning work, home, and on the go. We wanted a visual language that emotionally resonates across generations, works across platforms and devices, and echoes the kinetic nature of productivity today.

Our design solution was to decouple the letter and the symbol in the icons, essentially creating two panels (one for the letter and one for the symbol) that we can pair or separate. This allows us to maintain familiarity while still emphasizing simplicity inside the app.

Separating these into two panels also adds depth, which sparks opportunities in 3D contexts. Through this flexible system, we keep tradition alive while gently pushing the envelope.

Human-centered designs that emphasize content and reflect the speed of modern life

We all know modern life is faster and more connected: we’re living in it. Office supports this by making it fast and easy to express ideas, collaborate with others, and stay focused and in the flow. It’s why Office apps compose together, enabling users to open PowerPoint or Excel beside conversations in Teams or Outlook.

To reflect this in the icons, we removed a visual boundary: the traditional tool formatting. Whereas prior Office icons had a document outline for Microsoft Word and a spreadsheet outline for Excel, we now show lines of text for Word and individual cells for Excel. By focusing on the content rather than any specific format, these icons embody the collaborative nature of the apps they represent.

Similarly, we’ve changed the letter-to-symbol ratio. Traditionally, the letter occupied two-thirds of the icon, and the symbol took up one-third. We’ve changed this ratio to now emphasize the symbol because while the letter represents the tool itself, the symbol speaks more to people’s creations.

Being part of the design community

Our new icons will begin rolling out across platforms in the coming months, starting with mobile and web. They are the result of many iterations, a lot of research and testing, and plenty of late nights and weekends. They’re also part of an ongoing journey. As designers, we love the creative community’s ability to inspire each other and create momentum, so don’t hesitate to leave a comment below.

Go to Original Article
Author: Steve Clarke

Istio service mesh vies for lead in microservices market

Istio service mesh will capture attention as it integrates with Google Cloud Platform in 2019, but it hasn’t reached industry-standard status similar to its sister project, Kubernetes, just yet.

Istio, the service mesh technology created by IBM, Google and Lyft, reached version 1.0 this year. Istio service mesh integration with Google Cloud Platform will enter public beta tests in December 2018, according to Google, and become the default service mesh deployment option for GCP in the first quarter of 2019. Users will need to only check a box to include Istio deployment alongside Kubernetes clusters in GCP. In mid-2019, such checkbox deployment will come to the on-premises Google Kubernetes Engine product, and future versions of Istio integration with GCP will offer automated upgrades, similar to the ones GKE already makes available for Kubernetes clusters.

Istio service mesh is often associated with Kubernetes container orchestration, which also originated at Google, as container-based microservices have raised the profile of service mesh networking technology. Istio’s emergence alongside Kubernetes reflects maturity for enterprise container deployments.

Jay LymanJay Lyman

“There’s more talk about cloud-native architectures, and less about Kubernetes container orchestration specifically,” said Jay Lyman, analyst with 451 Research. “It’s early days for service mesh, but the need for it and the appeal of it will grow as container use grows.”

Service mesh is a network architecture with two main components: a data plane comprised of sidecar containers deployed on each container cluster host or Kubernetes pod, and a control plane that orchestrates the sidecar containers to manage traffic between microservices. The service mesh approach offers deeper and finer-grained performance and security monitoring insights into applications than network-or host-based monitoring software. Service mesh can also be used in polyglot application environments without the need to refactor network management tools for each application language.

Service mesh predates Istio; the term was first coined by Buoyant, makers of Linkerd, in mid-2016, to distinguish Linkerd’s approach from previous microservices network architectures that were deployed as libraries within specific applications. But Istio service mesh has gained momentum in 2018 because of its blue-chip backers and association with Kubernetes, and its rise to prominence will accelerate with the upcoming GCP integration.

“You certainly can’t declare Istio the winner in the market the way Kubernetes is, but it’s totally fair to call it the leader,” Lyman said.

HashiCorp Consul emerges as Istio service mesh frenemy

Istio service mesh faces serious contenders for dominance in the market for microservices networking technology, most notably HashiCorp Consul distributed service discovery and key/value store.

Consul’s rivalry with Istio as a microservices control plane is similar to other slippery relationships of coopetition in the market for open source cloud-native management tools — the two need not be mutually exclusive. For example, Istio requires a third-party service catalog from Kubernetes, Consul, Netflix Eureka or another source. Istio is Kubernetes-focused so far, though Google officials hinted that Istio service mesh support for VMs is on the project’s roadmap for 2019. For now, users can deploy Consul alongside Istio to manage noncontainer workloads. Consul encompasses multiple networking utilities, including domain name system, which can discover and connect to Kubernetes services even if the service mesh control plane is managed with Istio.

However, some early adopters of microservices prefer Consul as the overall control plane for service mesh. They also prefer the service discovery engine for Kubernetes over the default etcd key-value store.

Istio has good integration inside Kubernetes and claims it can manage workloads outside of Kubernetes, but there are a lot of ‘ifs’ and ‘buts.’
Rangan Prabhakaransenior financial software developer, Bloomberg LP

“We find Consul more distributed than etcd — individual Consul agents are the authority for health checks, rather than having to have a central authority,” said Rangan Prabhakaran, senior financial software developer at Bloomberg LP, the global finance, media and tech company based in New York. “That means we can perform more frequent health checks on tens of thousands of services with very high network traffic.”

Bloomberg plans to use Consul to manage service discovery on an infrastructure that consists of 20,000 nodes, some of which will be containers orchestrated with Kubernetes, while others will be bare-metal servers or VMs. It also plans to keep Consul as a centralized management interface for service mesh with the Envoy sidecar container, rather than Istio.

“Istio has good integration inside Kubernetes and claims it can manage workloads outside of Kubernetes, but there are a lot of ‘ifs’ and ‘buts,'” Prabhakaran said. “Istio also uses Consul for its service catalog, and we’re already familiar with Consul.”

HashiCorp increased its direct competition with Istio when it introduced Consul Connect in August. Consul Connect handles service discovery, authentication and authorization via transport layer security (TLS), and traffic management between multiple data centers for bare metal, VM and container-based workloads. The forthcoming GCP Istio service mesh integration will also include automated issuance of TLS certificates for Kubernetes clusters; multi-cloud network management is available for Kubernetes, managed via Istio, but it won’t be part of the Google Cloud integration initially.

Istio service mesh architecture
Istio’s service mesh architecture includes a control plane and a data plane that manage microservices network connections.

The addition of Consul Connect makes Consul a one-stop shop to enterprises for multi-cloud microservices networking that replaces etcd for Kubernetes service discovery, Istio service mesh, and separate inter-data-center ingress and egress service mesh projects such as Kong, VMware Heptio’s Gimbal and others.

Consul Connect will be a compelling product to watch as service mesh matures, HashiCorp customers said.

“Linkerd, which we’ve used so far, has a lot of moving parts, such as namerd and some glue code we wrote ourselves,” said Zack Angelo, director of platform engineering at BigCommerce, an e-commerce company based in Austin, Texas. “With Consul Connect, we could collapse all of that into one simplified set of tools.”

Connect could solve distributed state management, which is one of the most difficult problems in service mesh, Angelo said.

“Consul has mastered making that reliable within data centers,” he said. “We’ll keep a close eye on Consul Connect over the next six or seven months to see if it can do that between multiple data centers.”

Go to Original Article
Author:

Wanted – Mid range Graphics card

Discussion in ‘Desktop Computer Classifieds‘ started by Hughwp, Nov 21, 2018.

  1. Hughwp

    Hughwp

    Active Member

    Joined:
    Oct 4, 2007
    Messages:
    93
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Ratings:
    +5

    Some thing like a HD7770/GTX 750 or R7 or R9 270 thats no too power hungry and no longer than 10″
    WHY?
    Please quote price inc postage

    Location: Yeovil

    ______________________________________________________
    This message is automatically inserted in all classifieds forum threads.
    By replying to this thread you agree to abide by the trading rules detailed here.
    Please be advised, all buyers and sellers should satisfy themselves that the other party is genuine by providing the following via private conversation to each other after negotiations are complete and prior to dispatching goods and making payment:

    • Landline telephone number. Make a call to check out the area code and number are correct, too
    • Name and address including postcode
    • Valid e-mail address

    DO NOT proceed with a deal until you are completely satisfied with all details being correct. It’s in your best interest to check out these details yourself.

  2. Is a GTX 970 overkill? I have one I’m about to list due to upgrade. Looking for £90 plus postage.

  3. Pics.

    Attached Files:

  4. Hughwp

    Hughwp

    Active Member

    Joined:
    Oct 4, 2007
    Messages:
    93
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Ratings:
    +5

    Bit overkill for my needs only want it to resurrect an old PC and sell on so need to be cheap as not much market for s/h PCs
    Thanks anyway

  5. Devotee101

    Devotee101

    Active Member

    Joined:
    Mar 12, 2004
    Messages:
    157
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    21
    Location:
    Wigan
    Ratings:
    +7

    I’ll take it at £90 plus postage if you’re still selling @monkeyfist

Share This Page

Loading…

Go to Original Article
Author: