Category Archives: Cloud Servers
Watch Out, Amazon: Google Eyes the Cloud as its Next Frontier – PC Magazine
Google is a relative newcomer to enterprise cloud offerings, and it is using the success of its other products to win over companies that may be considering Amazon or Microsoft.
The cloud, to hear Google tell it, is all about attacking the corporate bureaucracy of Fortune 500 companies. Oh, and also improving internet security, going carbon neutral, creating magical creatures, catching Pokemon, and even automatically generating news stories.
In fact, at its Cloud Next conference in San Francisco on Wednesday, Google made a case for why pretty much any company connected to the internet should use its cloud computing offerings, which estimates suggest are significantly smaller than competitive services from Amazon and Microsoft, and dwarfed by the revenue of Google's own advertising business.
Perhaps Google's most convincing sales pitch is that its cloud offerings have already attracted many companies: marquee customers who have started using Google's cloud offerings over the past year include Verizon, HSBC, and eBay, which showed off a new cloud-powered Google Home app that can tell you how much money you can expect when you sell an old electronic gadget.
But Google's ability to move a bunch of data from a company's existing servers into the cloud isn't revolutionary: Microsoft and Amazon have been doing it for years, and Google really only started in earnest last September, when it created the new Google Cloud division. So its sales pitch also focuses on associating the company's traditional areas of expertiseespecially the popularity of its consumer products among millennialswith the cloud.
"It's Google's time to bring what we have to the enterprise," Diane Greene, senior vice president and head of Google Cloud, said at Cloud Next. "The young people so rejoice when G Suite gets rolled out," she said, referring to versions of Gmail and Google Docs for the enterprise, so why not move the rest of the IT department to Google's servers, too?
If companies decide to do that, they'll be able to access a slew of new tools that Greene and her colleagues announced today. Businesses that already run SAP applications will now be able to access them directly from the cloud platform. There's also a new engineering division that can provide round-the-clock support for challenges. (Google began to implement dedicated support last summer, when its teams helped Pokemon Go creator Niantic achieve "painless scaling," according to Greene).
Finally, there are a few new nifty mission-specific tools that some Cloud customers will find useful, including the ability to use Google's artificial intelligence to identify animals and objects in videos hosted on Google Cloud servers. That tool, called the Video Intelligence API, is essentially the same as the vision-recognition models that let you find cat videos on YouTube.
So why is Google building up its cloud business to compete with Amazon Web Services (which, as we know from its high-profile failure last week, hosts the images of a huge number of popular websites)? Google CEO Sundar Pichai (pictured above) pointed to the cloud as a natural fit given the company's focus on artificial intelligence and machine learning.
"Machine learning finds previously hidden patterns in data," Pichai said. Greene, meanwhile, pointed to the untapped potential of data that's trapped in old fashioned databases and servers. "Ninety-five percent of the world's data is not in the cloud," she told the Wall Street Journal.
But Eric Schmidt, executive chairman of Google's parent company Alphabet, offered a simpler explanation. He suggested at Cloud Next that Google is embarking on a journey to the cloud simply because it can. "We have the means, commitment, and money" to pull off a new cloud platform for everyone who needs it, he said.
Tom is PCMag's San Francisco-based news reporter. He got his start in technology journalism by reviewing the latest hard drives, keyboards, and much more for PCMag's sister site, Computer Shopper. As a freelancer, he's written on topics as diverse as Borneo's rain forests, Middle Eastern airlines, and big data's role in presidential elections. A graduate of Middlebury College, Tom also has a master's journalism degree from New York University. Follow him on Twitter @branttom. More
See the original post here:
Watch Out, Amazon: Google Eyes the Cloud as its Next Frontier - PC Magazine
Microsoft, Facebook Build Dualing Open Standard GPU Servers for Cloud – TOP500 News
It was only a matter of time until someone came up with an Open Compute Project (OCP) design for a GPU-only accelerator box for the datacenter. That time has come.
In this case though, it was two someones: Microsoft and Facebook. This week at the Open Compute Summit in Santa Clara, California, both hyperscalers announced different OCP designs for putting eight of NVIDIAs Tesla P100 GPUs into a single chassis. Both fill the role of a GPU expansion box that can be paired with CPU-based servers in need of compute acceleration. The idea is to disaggregate the GPUs and CPUs in cloud environments so that users may flexibly mix these processors in different ratios, depending upon the demands of the particular workload.
The principle application target is machine learning, one of the P100s major areas of expertise. An eight-GPU configuration of these devices will yield over 80 teraflops at single precision and over 160 teraflops at half precision.
Source: Microsoft
Microsofts OCP contribution is known as HGX-1. Its principle innovation is that it can dynamically serve up as many GPUs to a CPU-based host as it may need well, up to eight, at least. It does this via four PCIe switches, an internal NVLink mesh network, plus a fabric manager to route the data through the appropriate connections. Up to four of the HGX-1 expansion boxes can be glued together for a total of 32 GPUs. Ingrasys, a Foxconn subsidiary will be the initial manufacturer of the HGX-1 chassis.
The Facebook version, which is called Big Basin, looks quite similar. Again, P100 devices are glued together vial an internal mesh, which they describe as similar to the design of the DGX-1, NVIDIAs in-house server designed for AI research. A CPU server can be connected to the Big Basin chassis via one or more PCIe cable. Quanta Cloud Technology will initially manufacture the Big Basin servers.
Source: Facebook
Facebook said they were able to achieve a 100 percent performance improvement on ResNet50, an image classification model, using Big Basin, compared to its older Big Sur server, which uses the Maxwell-generation Tesla M40 GPUs. Besides image classification, Facebook will use the new boxes for other sorts deep learning training, such as text translation, speech recognition, and video classification, to name a few.
In Microsofts case, the HGX-1 appears to be the first of multiple OCP designs that will fall under its Project Olympus initiative, which the company unveiled last October. Essentially, Project Olympus is a related set of OCP hardware building blocks for cloud hardware. Although HGX-1 is suitable for many compute-intensive workloads, Microsoft is promoting it for artificial intelligence work, calling it the Project Olympus hyperscale GPU accelerator chassis for AI, according to a blog posted by Azure Hardware Infrastructure GM Kushagra Vaid.
Vaid also set the stage for what will probably become other Project Olympus OCP designs, hinting at future platforms that will include the upcoming Intel Skylake Xeon and AMD Naples processors. He also left open the possibility that Intel FPGAs or Nervana accelerators could work their way into some of these designs.
In addition, Vail brought up the possibility of a ARM-based OCP server via the companys engagement with chipmaker Cavium. The software maker has already announced its using Qualcomms new ARM chip, the Centriq 2400, in Azure instances. Clearly, Microsoft is keeping its cloud options open.
Read the rest here:
Microsoft, Facebook Build Dualing Open Standard GPU Servers for Cloud - TOP500 News
AMD’s 32-core, Zen-based Naples chip aims to break Intel’s server dominance – PCWorld
The outspoken Forrest Norrod has never shied away from challenges. Previously, as a server chief at Dell, he helped the company's data-center hardware business flourish, and he loved experimenting with new types of servers.
He has a new challenge as AMD's server chief: to bring back the glory days of chipmaker's server business, which is now in tatters. A mega-chip called Naples, which has 32 cores and is based on the Zen architecture, will be the first test of AMD's return to the server market.
The Naples chip will ship to server makers in the second quarter of this year. The benchmarks of Naples are competitive with Intel's chips in head-to-head comparisons, said Norrod, senior vice president and general manager of AMD's Enterprise, Embedded, and Semi-Custom Business Group.
AMD's return will add much-needed competition to the server chip market, which Intel dominates. Intel has more than a 90 percent market share, and AMD's goal is to steadily siphon off customers.
Customers may welcome AMD's server chips because Intel's chips are priced high, with the most expensive chip selling for $8,898. Lower priced AMD chips could give customers bargaining power.
"If we look at how they price their consumer products, it stands to reason that the versions of Naples will also undercut Intel's pricing," said Nathan Brookwood, principal analyst at Insight 64.
An AMD discount is already happening with the company's new Ryzen desktop chips, which are significantly cheaper than Intel's gaming chips.
The Naples server chipsare based on the x86 architecture, but they don't have an official name yet. AMD was once a legitimate threat to Intel, but a series of missteps killed its server business. The decline started with the heavily criticized Bulldozer architecture, and the company later bet its server future on ARM chips, but slow demand knocked the company out of servers.
Naples won't be the only server chip based on Zen. More chips will be revealed as the server chip launch comes closer, Norrod said. Not everyone will buy a 32-core chip, so AMD may release server chips with fewer cores.
The response to Zen has so far been "delightful," Norrod said.
Zen server chips are targeted at single- and two-socket servers, which are used for general-purpose and cloud applications. A two-socket system could have eight memory channels, which could help with in-memory applications like databases, which require more bandwidth.
Naples will have more I/O and memory capacity than comparable Intel chips, Narrod said. Those features should help with machine learning, especially when the chip is paired with co-processors like GPUs. AMD hopes to pair Zen server chips with its upcoming Vega GPU, which will be tuned for machine learning.
The high-bandwidth fabric linking two sockets in the chips is unique, which makes them ideal for two-socket servers, Brookwood said.
AMD has already licensed its new server architecture to THATIC (Tianjin Haiguang Advanced Technology Investment Co. Ltd.), a joint venture in China that is making surrogate Zen chips for the local market. That doesn't mean AMD will hold its Naples chips from the China market, Norrod said.
The Naples chip will take on Intel's Skylake server chips, which are scheduled to be used by Google in its cloud servers. One area where the AMD chips will fall short is in high-performance applications, where Intel chips could excel. Intel's Skylake server chips will have AVX-512 to run vectorized applications, while AMD's chips have only AVX-128.
There are some algorithms that can benefit from AVX-512, and Intel has been in the high-performance computing area longer than AMD, Brookwood said.
"Intel knows where those codes are buried," while AMD is still trying to discover the market, Brookwood said.
There are some applications that will run better on Intel's chips, Norrod said. But AMD's Naples has its own unique features that could help the company come firing back into the server market, he said.
In the end, if AMD's chip is competitive on performance, even if slightly slower, customers will consider using it.
"If you have comparable performance, then pricing's a big deal," Brookwood said.
More here:
AMD's 32-core, Zen-based Naples chip aims to break Intel's server dominance - PCWorld
Securing applications in the public cloud – Computerworld
I have written on the topic of cloud-induced transformation of IT in the past. Adapting IT audit and monitoring processes to cloud infrastructure is one of the challenges I come across when it comes to cloud rollouts.
In a 1990s-era data center, everything revolves around hardware and virtual machines. Big, monolithic applications are installed and run on servers. Servers themselves run in the private subnet (secure) or public (DMZ), and they have various security agent software installed to monitor and log everything that goes in and out of these machines.
It is easy to think of public cloud (such as AWS) as a managed hosting service or collocation. However, this is only a fraction of the services public clouds offer. Among services provided by large public cloud providers like AWS or Azure, there is storage, queuing, machine learning, container hosting, database engines and much more.
There are other application platform services out there in the public cloud. Consider the REST API services offered by Microsoft Office 365, Salesforce, Google and LinkedIn: None of them involve any virtual servers at all.
The services that do create virtual server instances do so in an entirely automated fashion. Consider AWS RDS, for example, which under the covers spins up virtual servers that run the DBMS software. Likewise, the AWS Elastic Container Service, Elastic MapReduce and Kenesis create fully managed EC2 instances. It would defeat the purpose of using the public cloud to try and manage these servers on your own.
A correctly built cloud-first application server is transient. This server scales automatically with the workload, wakes up when needed and goes to sleep when not. Developers don't log on to these servers directly. It does not make sense to monitor these servers in a traditional way.
It is possible to build a modern app without using any backend server component at all. Useful applications can be put together using API mashups, and APIs offered by social media and cloud providers.
These applications require no custom backend code at all. Any backend code that is needed can be built using something like AWS Lambda functions. The APIs and functions execute on servers that IT has very little or no control over.
Each cloud provider provides tools for security monitoring, logging and audits. If a cloud provider does not offer any, then perhaps the choice should be re-examined.
While Azure has similar tools and procedures, I will focus on AWS, since it is the one I am most familiar with. IT teams should review the AWS security audit guide.
At the infrastructure level, consider AWS Elastic Load Balancer log collection. Operations teams can collect and analyze network traffic using VPC flow logs.
There are a few more options for additional security. Most importantly, any additional tiers that implement security must meet or exceed application availability and scalability requirements. Before configuring any custom in-line gateway or forward proxy, I would still recommend first exhausting native AWS resources.
Before NAT Gateways became available, AWS recommended configuring a NAT instance. NAT instances required additional administrative and DevOps effort. When NAT Gateways as a service became available, it was an instant hit with DevOps teams. Unlike a NAT EC2 instance, the AWS Gateway service offers better availability and more bandwidth -- and it works well for most use cases. Likewise, AWS should offer a managed forward proxy as a service.
The real challenge, however, is ensuring that the EC2 instances in the private subnet can only talk to approved external services -- and that includes AWS APIs such as Dynamo, Kinesis, S3 and SQS. The purpose of VPC endpoints is to allow applications to communicate with AWS services without going through the public internet.
Unfortunately, AWS only offers an S3 endpoint. That severely limits AWS services that can be used from the private subnet without having to jump through hoops. Creating VPC endpoints for all AWS services should be at the top of the list for AWS.
This article is published as part of the IDG Contributor Network. Want to Join?
Read the original:
Securing applications in the public cloud - Computerworld
Deploy Dedicated Servers Instantly with ServerMania Instant Bare … – PR Web (press release)
Clients had to choose between optimum power and immediate availability with Instant Bare-Metal they can have both.
Toronto, ON (PRWEB) March 08, 2017
ServerMania, a leading provider of cloud and dedicated server hosting, has announced the introduction of Instant Bare-Metal, enterprise-grade bare metal servers that can be deployed in seconds.
Instant Bare-Metal combine the instant availability of cloud servers with the raw performance of bare metal. Instant Bare-Metal can be deployed immediately, and are the perfect option for infrastructure hosting clients who need immediate access to the most powerful servers.
ServerMania has selected its most popular dedicated server plans and made them available at the click of a button. The servers are ready and waiting in their racks, and with minimal configuration, can be lit up and made available to clients.
For workloads that demand optimal processing power, IO, and memory allocation, you cant beat a dedicated server, explained ServerMania CEO, Kevin Blanchard, But dedicated servers usually take time to configure and deploy. Clients had to choose between optimum power and immediate availability with Instant Bare-Metal they can have both.
Instant Bare-Metal are a perfect fit for a wide range of use cases. Software-as-a-Service businesses with I/O intensive operations depend on the ability to scale quickly. With Instant Bare-Metal, they can immediately deploy dedicated servers capable of supporting the most demanding workloads. Other scenarios that are a perfect fit for Instant Bare-Metal include the fintech sector, where immediate access to high-performance computing infrastructure is essential; and real-time big data analytics workloads, which benefit from access to the fastest I/O and uncompromised processing power.
ServerMania offers a comprehensive range of bare metal and cloud hosting options, including dedicated servers, public and private cloud hosting, and colocation.
###
About ServerMania: Since it was founded in 2002, ServerMania has always strived to provide its clients with enterprise-level service at an unbeatable cost. ServerMania offers a wide range of fully customizable dedicated, hybrid, cloud, VPS and colocation hosting services. All ServerMania clients enjoy a 100% uptime SLA and are assisted by a 24/7 rapid response team one with some of the best response times in the industry. ServerMania also carries out regular surveys to ensure complete customer satisfaction and care. For more information, visit http://www.servermania.com.
Share article on social media or email:
Follow this link:
Deploy Dedicated Servers Instantly with ServerMania Instant Bare ... - PR Web (press release)
Nextcloud releases security scanner to test deployments – iTWire
A survey of private cloud deployments has led the open source cloud provider Nextcloud to issue a security scanner in order that users can find out the state of their security and update as needed.
The company developed the scanner after examining the state of cloud deployments using the shodan.io search engine. This uses a very basic scan that provides the version number of the cloud software in use so that its security state can be ascertained.
After realising that many were lacking updates that rendered them vulnerable, Nextcloud decided to warn administrators of these instances.
The company said it did not feel comfortable noting that political parties, hospitals, universities, large corporations and governments were figuring in a list of insecure servers.
"They decided to reach out to users with a personal warning, including the results of the scan. We made sure the security scan would not expose any private data, using unique IDs instead of URLs to present them the results and we kept as quiet as possible on our communication channels about this matter," Nextcloud communications manager Jos Poortvliet said.
The results were good, with 5% upgrading within the first 10 days of being notified.
But even so, Poortvliet said, the company;s estimates were that there were at least 100,000 private cloud servers whose owners were unaware that they were sitting ducks for attacks.
As a result, the Nextcloud update tool was rewritten to make it easier to use from the command line in version 11. From version 12, apps would not be disabled during a security update making the process less intrusive.
"Our ultimate goal is to make updates so seamless they can be done fully automatic without any administrator involvement or downtime. At this moment, we have achieved this on the Nextcloud Box, using Canonicals Snap technology which automates updates entirely," Poortvliet said.
Read the original:
Nextcloud releases security scanner to test deployments - iTWire
Amazon explains big AWS outage, says employee error took servers offline, promises changes – GeekWire
(Shutterstock Image).
Amazon has released an explanation of the events that caused the big outage of its Simple Storage Service Tuesday, also known as S3, crippling significant portions of the web for several hours.
Amazon said the S3 team was working on an issue that was slowing down its billing system. Heres what happened, according to Amazon, at 9:37 a.m. Pacific, starting the outage: an authorized S3 team member using an established playbook executed a command which was intended to remove a small number of servers for one of the S3 subsystems that is used by the S3 billing process. Unfortunately, one of the inputs to the command was entered incorrectly and a larger set of servers was removed than intended.
Those servers affected other S3 subsystems, one of which was responsible for all metadata and location information in the Northern Virginia data centers. Amazon had to restart these systems and complete safety checks, a process that took several hours.In the interim, it became impossible to complete network requests with these servers. Other AWS services that relied on S3 for storage were also affected.
About three hours after the issues began, parts of S3 started to function again. By about 1:50 p.m. Pacific, all S3 systems were back to normal. Amazon said it has not had to fully reboot these S3 systems for several years, and the program has grown extensively since then, causing the restart to take longer than expected.
Amazon said it is making changesas a result of this event, promising to speed up recovery time of S3 systems. The companyalso created new safeguards to ensure that teams dont take too much server capacity offline when working on maintenance issues like the S3 billing system slowdown.
Amazon is also making changes to its service health dashboard, which is designed to track AWS issues. The outage knocked out the service health dashboard for several hours, and AWS had to distribute updates via its Twitter account and by programming in text at the top of the page. In the message, Amazon said it made a change tospread that site over multiple AWS regions.
Amazon concluded its explanation with this message:
Finally, we want to apologize for the impact this event caused for our customers. While we are proud of our long track record of availability with Amazon S3, we know how critical this service is to our customers, their applications and end users, and their businesses. We will do everything we can to learn from this event and use it to improve our availability even further.
Several observers surveyed by GeekWirepointed to the need for redundancy in cloud storage asa key takeaway from the outage.Redundancy in this case can mean spreading data across multiple regions, so that an outage in one area doesnt cripple an entire site, or using multiple cloud providers.
Anand Hariharan, vice president of products for Mountainview, Calif.-based Webscale Networks noted that Amazons retail website didnt go down duringthe outage Tuesday because it didnt put all its eggs in one cloud basket.
As AWS incredibly disruptive outage this week showed, every major public cloud provider has experienced or will experience downtime. In fact, more and more of our customers particularly those running e-commerce businesses recognize that they cant just rely on one cloud provider, or one region. Amazon themselves stayed live and fast because they do exactly this spread their infrastructure across multiple regions. Hours and really just minutes of downtime are a lifetime for businesses. Downtime costs not only revenue, but brand reputation and consumer trust, so companies need to consider their multi-region/multi-cloud strategies today.
The internet reacted in a pretty jovial manner to the outage Tuesday, with many taking the outage as a chance for adigital snow day. Amazons explanation of the outage earned praise from some for the companys transparency and scorn from others.
Microsoft said to cut purchases of HPE servers for cloud service – Information Management
(Bloomberg) -- Hewlett Packard Enterprise Co. is losing business from Microsoft Corp., one of the worlds largest users of servers, the latest sign of trouble for the pioneering computer maker as it struggles with the rise of cloud services, people familiar with the matter said.
Hewlett Packard Enterprise Chief Executive Officer Meg Whitman said last week her company saw "significantly lower demand" for servers from a tier-1 service provider, but without identifying the customer. Tier-1 service providers are typically major cloud and telecom companies.
The softer demand came from Microsoft, the people said, as the software giant pushes for lower prices from hardware providers to help it efficiently expand its public cloud service and keep up with rivals Amazon.com Inc. and Alphabet Inc.s Google. Spokeswomen for Microsoft and HPE declined to comment.
Late last year, the Redmond, Washington-based company unveiled a new in-house cloud server design that it will require hardware vendors to follow. This forces HPE and rival Dell Technologies Inc. to compete against lower-cost generic, commodity manufacturers. Already, Microsoft has been using less-expensive gear for its data centers and the new design is set to be fully implemented later this year.
"We will continue to meet customer demand by expanding data center capacity while driving efficiencies through new technologies," Microsoft Chief Financial Officer Amy Hood said on a call with analysts in January when it announced earnings.
Microsofts Azure public cloud business reported a 93 percent revenue surge in the final quarter of 2016 as more businesses opted for the flexibility and ease of accessing computing power and storage over a network instead of building their own data centers.
HPE has been a leading seller of servers that go in these corporate data centers. But the shift to the public cloud means businesses dont need to buy their own servers anymore. Selling to the big cloud providers is harder, either because they demand more volume discounts, or increasingly they design their own cheaper servers.
"Within Tier 1 we had a much lower demand from a single large customer," Whitman said on the call, noting that these types of deals arent as profitable as other parts of the server business. "The Tier 1 business is very competitive, and well see what happens there."
HPE last week cut its adjusted profit forecast for the current fiscal year and reported sales that missed analysts projections for the third consecutive quarter.
Visit link:
Microsoft said to cut purchases of HPE servers for cloud service - Information Management
Massive AWS Cloud Server Outage Was Caused by an Incorrect Debugging Command – 1redDrop (blog)
When youre a company thats disrupting nearly everything it touches, anything you do is reported almost instantaneously in the media. And that attention is amplified when you take down the websites of some of the worlds most known techbrands for 5 hours or so. Thats what happened to Amazon Web Services on Tuesday, February 28, 2017.
The effects of the outage have been discussed to shreds, but it now turns out that a simple mistakein a command given during a debugging run of its billing system was the root cause of the outage.
The wrong command didnt actually cause the outage, but it took down the wrong S3 servers, forcing a full restart. Considering the scale at which AWS servers operate, that took a few hours hence, the outage.
The fact that a wrong command can execute the wrong operation is understandable. After all, thats where human error plays a role. The interesting thing is that Amazon is now going to make changes to the system so incorrect commands are not able to trigger an outage like this one.
We dont know what those changes are other than the fact that capacity removal is being capped (the allowable limit was too high) and recovery times are greatly reduced, but will that prevent another outage from ever happening? Hardly. Its just one hole thats being plugged.
What we need to understand is that any technology can be broken. IBM says that cloud can be made more secure than the static security layers typically found in traditional infrastructure, but whos going to account for human error, as was clearly the case in the AWS outage?
Thanks for reading our work! We invite you to check out ourEssentials of CloudComputingpage, which covers the basics of cloud computing, itscomponents, variousdeployment models, historical, current andforecast datafor the cloud computing industry, and evenaglossaryof cloud computing terms.
Source
The rest is here:
Massive AWS Cloud Server Outage Was Caused by an Incorrect Debugging Command - 1redDrop (blog)
Amazon’s web servers are down and it’s causing trouble across the internet – The Verge
Amazons web hosting services are among the most widely used out there, which means that when Amazons servers goes down, a lot of things go down with them. That appears to be happening today, with Amazon reporting high error rates in one region of its S3 web services, and a number of services going offline because of it.
Trello, Quora, IFTTT, and Splitwise all appear to be offline, as are websites built with the site-creation service Wix; GroupMe seems to be unable to load assets (The Verges own image system, which relies on Amazon, is also down); and Alexa is struggling to stay online, too. Nests app was unable to connect to thermostats and other devices for a period of time as well.
Isitdownrightnow.com also appears to be down as a result of the outage.
Amazon has suffered brief outages before that have knocked offline services including Instagram, Vine, and IMDb. There dont appear to be any truly huge names impacted by this outage so far, but as always, its effects are widespread due to just how many services especially smaller ones rely on Amazon.
Theres no estimate on when service will be restored, but Amazon says it is actively working on remediating the issue.
Read the original here:
Amazon's web servers are down and it's causing trouble across the internet - The Verge