According to the company, Draft was developed to address the complexity and constraints the development community was facing when it came to working with Kubernetes.

“Application containers have skyrocketed in popularity over the last few years. In recent months, Kubernetes has emerged as a popular solution for orchestrating these containers. While many turn to Kubernetes for its extensible architecture and vibrant open-source community, some still view Kubernetes as too difficult to use,” Gabe Monroy, lead PM for containers on Microsoft Azure, wrote in a blog post.

Draft allows developers to start working on container-based apps without having to install Kubernetes or use Docker, Monroy explained.

“This simple yet flexible user experience for developers is inspired by PaaS systems like Deis and Cloud Foundry which support the concept of buildpacks. However, Draft differs from buildpack-oriented PaaS systems because it writes out build and deployment configuration into the source tree, making it trivial to construct continuous integration (CI) pipelines that can bring these containers all the way to production,” Monroy wrote.

In addition, Microsoft today announced the public preview of Azure Site Recovery. The solution enables businesses to create a disaster recovery strategy for IaaS based apps on Azure. It is offered “as-a-Service,” provides a simplified user experience, features application-aware recovery, and non-disruptive DR drills.

More information is available here.  

The post CoreOS Fest: Microsoft introduces Draft for app development on Kubernetes appeared first on SD Times.

Part of this evolution is that companies are now using parts of OpenStack, instead of treating the platform like a monolithic system. And according to Edward Snowden, who live-streamed in at the conference from Russia, it’s today’s OpenStack programmers that could greatly impact the future of cloud infrastructure and how it’s built.

Besides Snowden’s appearance, vendors released updates to their solutions and platforms. SD Times gathered some major highlights from the summit here:

Red Hat adds open source to MOC
Red Hat had a number of announcements last week at its own conference, but the company saved some announcements for this year’s OpenStack Summit. Red Hat announced a new collaborative open cloud initiative built on Red Hat’s OpenStack Platform and Red Hat Ceph storage. The Massachusetts Open Cloud (MOC) is now dedicated to publicly-available cloud computing resources for running Big Data workloads, according to the company.

MOC will provide researchers and businesses with a common, cloud-based infrastructure so they can gain insight from their volumes of data, said Red Hat. In addition to MOC, Red Hat announced its latest version of its OpenStack platform, which comes with enhanced support for upgrades with compassable roles, and new networking capabilities.

Platform9 joins OpenStack marketplace for private cloud
Platform9, a cloud infrastructure company, announced that it is now a part of OpenStack’s marketplace for remotely managed private cloud. The OpenStack Marketplace lets users research services from the OpenStack ecosystem, based on criteria like tests, API versions, hardware requirements, and more.

Rackspace, Dell EMC partnering for OpenStack
A new partnership was revealed at this year’s summit. Rackspace and Dell EMC are partnering up in order to delivery OpenStack clouds. Now, Rackspace will contribute its cloud expertise with Dell EMC’s computer and storage solutions so customers can create private cloud-as-a-service offerings.

Inspur In-Cloud OS 5.0 release
Inspur, an OpenStack solution provider, debuted its In-Cloud OS 5.0 smart cloud operating system, designed for cloud data centers and cloud-native applications. The system is built on the OpenStack architecture, and it provides users with functionality, availability, security and toolchain for the cloud service directory. In-Cloud OS 5.0 also supports deployment on Docker.

EasyStack launches ESContainer
The OpenStack platform provider EasyStack officially launched its Kubernetes-based enterprise container clustering solution, ESContainer. It provides auto-deployment, user interfaces, a marketplace for applications, monitoring, and convergence between EasyStack’s enterprise-leek OpenStack platform, ESCloud.

The ESContainer is designed for application based deployment and it allows applications to be deployed, published and managed on the same platform. For large scale container cluster deployment, the solution also provides monitoring and log journaling for both development and operations teams, said the company.

The post OpenStack Summit: Private cloud, infrastructure is changing appeared first on SD Times.

This new release of Tectonic includes support for setting custom worker subnets inside the installer. Also new is support for storage configuration during setup. This release eliminates the need for elastic IP addresses.

(Related: Navigating the container ecosystem with Kubernetes 1.5)

Tectonic 1.5 also includes a basic YAML editor, allowing users to edit configurations and settings right within the Tectonic control panels. Another major focus was preparing for a smoother upgrade path in future releases.

For this version, users must upgrade to Kubernetes 1.5.2, but that upgrade path seems to be only available from Kubernetes 1.4.7 or 1.5.1. Future releases of Tectonic should be able to handle all upgrades, provided users are already on Kubernetes 1.5.2.

The primary use of Tectonic, after all, is to automatically patch and maintain the infrastructure surrounding and within a Kubernetes cluster. CoreOS dubs this “self-driving” capability.

Alex Polvi, CEO of CoreOS, wrote, “Kubernetes is on its way to becoming the standard for cloud-native plumbing. With self-driving Kubernetes, organizations can ensure this critical infrastructure layer is always up to date with the latest patches and features, and our customers can enjoy the benefits of Continuous Delivery of upstream innovation.”

The post CoreOS ships Tectonic 1.5 container-based infrastructure platform appeared first on SD Times.

Many of the biggest software IPOs to have in the last few years have been SaaS-related. However, there are many failed examples as well. Many developers are struggling to understand the core essence of the SaaS delivery platform and fail to design robust, scalable and secure architectures for SaaS.

This article is based on the lessons learned from developing our flagship product, Clintra, a cloud-based business–management system. It highlights various aspects you as a SaaS developer must focus on to make your SaaS offering profitable for the long run. It is an exciting (but at the same time challenging) journey.

Why should you take notice?
Worldwide spending on public cloud services will grow at a 19.4% compound annual growth rate (CAGR) from nearly US$70 billion in 2015 to more than $141 billion in 2019. According to an IDC report, “Worldwide Semiannual Public Cloud Services Spending Guide.” IDC predicts SaaS will remain the dominant cloud computing type, capturing more than two-thirds of all public cloud spending through most of the forecasted period. Worldwide spending on infrastructure-as-a-service (IaaS) and platform-as-a-service (PaaS) will grow at a faster rate than SaaS with five-year CAGRs of 27% and 30.6%, respectively.

Gartner predicts that application software sales will grow by 7.5% this year, and will exceed $200 billion by 2019, driven by widespread adoption of the cloud-based SaaS model.

So are you ready to build your next unicorn company and join the billionaires club? If so, let’s first focus on the important aspects you need to master before you turn your existing software or new idea into SaaS.

Three important factors for building secure, scalable and profitable applications for the cloud
SaaS architecture is different than traditional software architecture. Whether you want to convert your traditional software to SaaS or build it from scratch, you must pay close attention to some of the important factors of the architecture, such as security, performance, scalability and availability.

Security, the all-important beast: Security is the most important factor of SaaS, and every CIO will pay close attention to it. Almost all of our clients asked us if our software is secure and whether or not their data will be safely stored.

With many companies affected by hackers who steal valuable company and customer information, many customers have become more cautious of using SaaS applications. Thus, it is safe to say that if your online service ends up in the news for all the wrong reasons, there is a high probability that there will be a fall in the number of active users of that service—at least immediately.

There are multiple security measures you can take. You need to make sure you have SSL installed on the server, which enables data to be transferred securely. For user authentication and authorization, you can choose from multiple options, like two-factor authentication, IP blocking and MAC address blocking. There are some development practices and design patterns you can use, which will allow to secure your APIs.

Performance and availability, and tips on how to ride this mammoth: Technological innovation has resulted in people taking many features for granted. Users have an ever-increasing set of criteria that must be met in order for them to consider an online service to be exceptional.

This has proven to be a challenge for many vying to provide the highest quality web-based SaaS. If you have used one yourself, you might have found some to be glitchy, slow and downright complicated to use, which users do not enjoy. Thus, if you want to ensure that your online software solution remains the favored option among your users, make sure it performs according to their liking.

In traditional software there will be only one user logged into the system at a time, and the performance of the application depends solely on the user computer’s resources. For SaaS applications, it is different as the processing load is shifted to the server-side, and the user’s computer is acting as a dumb client.

Also, there will be multiple users connecting from different parts of the world at the same time. This puts a heavy load on the server resources, so you want to make sure your server can handle this properly. One way to deal with this is to load balance your services. You can add multiple server nodes, which can connect to a single load balancer. This way you can distribute the load across multiple servers.

There are different designs you can use for load balancing your server, such as round robin, multicast/broadcast, etc. You will also need to focus on failover strategies. For example, you should be prepared to answer the following questions: What will happen if the load balancer fails? Should you load balance your load balancer? What kind of impact will users have if one application node goes down? What will happen if the master database node goes down? What are the points you will include in your disaster recovery plan?

Data backup and recovery, or how to catch the fly: Your disaster recovery plan must focus on data backup and recovery.

This aspect has some legal issues associated with it as well. Depending on your domain area, you may need to make sure you take proper measures to secure your data. For example, if you have a medical SaaS application, then your data backup and recovery plan will look significantly different than an ERP SaaS application. Also, you need to make sure you create multiple recovery strategies. In case one strategy fails, you should have another strategy available.

For data backup, you need to make sure you keep database backups as well as document backups. For database backups, you can set auto-replication on a master database to a read-only slave database server. This will enable you to use this slave server to replace your master database server in case of failure.

For document storage, you need to replicate the stored document from your server to an external storage backup in case you lose your primary storage. One way to do it is use a storage service from AWS, Google Cloud Storage, or some other storage provider. You can setup rsync on the storage folder, which will automatically replicate the documents to this secondary storage.

Three soft factors for making SaaS usable
Once you took care of the above important factors, you must make sure your application is usable. Many SaaS platforms do not do well because they only focus on getting the above three factors right but completely miss out on usability.

There are three soft factors you should look at in order to reap the rewards from your SaaS offering:

Quality of service: Quality is king, especially keeping in mind the number of services that can be used to perform the same task. If a user doesn’t like an online software service, they will graduate to using one provided by your competitor. What I mean by “quality” is concerned with the interface of the service, how stable it is, and how fast it opens, among other factors.

Availability: The mistake that some companies make when they launch a SaaS is that they limit the number of people who can use that service. This is done by only making the service available in specific cities, countries and regions. Even though it is understandable as to why they opt to do this (to test the waters instead of risking it all), it should be noted that this might cause them to lose out on potential business and provide an opportunity for competitors to step in and fill the void.

Ease of use: Online-based software should be easy to use. The main purpose behind providing online software services is to provide convenience to the users. For this reason, online service providers should make sure they stick to a simple format of the software, one that is the same as the one that is offered in the offline version. More often than not, keeping it simple is the best option.

The trade-offs between SaaS performance, security and usability
Sadly, you can’t have everything you wish for with a SaaS feature. You have to look for a balance between performance, security and usability. It is a triangle. The more you stretch on security the less usable the system is. The same is true for performance; the more you focus on performance, the chances are you will create a few security holes in the system. The more you focus on usability, the security and performance get affected.

For example, in order to increase security, you added two-factor authentication. Now the user has to carry an external device with them all the time to make sure they get the valid code for login. This affects usability, and from the performance side the server has to perform additional authentication to validate the entered code.

However, you can also provide options so that your end user can decide whether they want to focus on security or usability. For example, for Clintra, since it is a very generic application that can be used by any industry, we decided to include all the authentication measures, like two-factor authentication, IP blocking, MAC blocking, etc., and allow our clients to choose which type of security they would like to enable.

This allows them to choose between security and usability, so you as a SaaS provider do not have to force your clients one way or the other.

How to choose the best cloud infrastructure for your application
Infrastructure plays an important role in making your SaaS platform profitable, scalable and usable. As such, the world is moving from an owning to a leasing mindset. Leasing offers a lot of benefits as there is no huge upfront cost, and you can start off with a small investment.

You should design your SaaS architecture in a way that it allows you to use a PaaS and IaaS in combination. The PaaS model makes the coding and programming of web-based software easy by providing developers with the tools to make the service better in a number of aspects. There are many great PaaS services available that offer low-level services like media server, Auth Server, preconfigured application servers, etc., such as Salesforce.com.

These services will save you considerable expense because you don’t need an upfront investment to hire someone to set up your servers for you and then pay them yearly to manage them.

PaaS comes with support so you can get peace of mind and save money. The same is true with IaaS, which provides the combination of software and hardware that is used to provide the service. It focuses on the operational aspect and proper functioning of the web-based services. In simpler terms, it is a combination of servers, networks, storage and operating systems that is used to effectively deliver the service.

You should focus on using a mix of PaaS and IaaS platforms. This will enable you to do rapid development and improve time to market for your SaaS-based applications.

Why are developers struggling?
SaaS is growing, and many new companies are reaping the profits from the huge market for SaaS-based products. However, older developers and companies are feeling left behind.

Whenever there is a paradigm shift, there is an opportunity for new ideas to overtake old, established ideas. This is exactly what is happening with SaaS. It totally changes the rules of the game, and while older companies struggle to change and adapt, newer companies made for the future are leapfrogging over them.

A new revenue model
Revenue models for software companies were simple. You usually sold your product in an annual license and added in a monthly payment for support in some cases.

SaaS introduces a new revenue model, which is different. Instead of buying your product, people buy a subscription to use your services.

SaaS often ties customers into contracts for a certain period of time, say 12 months. This makes your revenue forecast much more predictable, which investors like to see.

Software vendors no longer collect large sums upfront, but fees are spread out over time. This allows them to scale the offering to serve many more customers at the same time, and also to deliver the product more rapidly to their customers all over the world.

Customers also get a lot of customization options, which further complicate financial projections and models.

Companies have to figure out how to generate revenue in a way that is completely different from anything they have done before. It is hard enough to rebrand a company—it is much harder to rethink the company’s business model.

The need to rebuild the software
Most established companies have enough people to update their software and make new versions of it. In order to adapt traditional software applications to SaaS, however, companies have to completely rewrite their software from scratch in many cases. New companies can simply invest all their development resources in creating new software in a SaaS model. Older companies that are dependent on legacy software need to support their existing software while creating a separate SaaS product as well.

Companies also have to completely rethink how their software works in order to turn it into a SaaS product. There are software products that have been on the market for two decades, and they were always built with the assumption that they will be installed on computers. But that isn’t true anymore.

The SaaS platform has become increasingly popular for many reasons. Generally, customers find the subscription-based solution more flexible in terms of their needs. It also makes their products and services more affordable for them in the long term. By making the switch to the SaaS paradigm, developers can capture a large segment of the target market, and earn greater revenues.

Here are the some of the things that can help make the transition to a SaaS platform successful and profitable for a developer:

• Deliver unique and outstanding products and services that not only satisfy, but delight customers.
• Keep the product offering simple. Complex features seem intimidating to new customers, causing them to look elsewhere.
• Dedicate resources such as sales, services, and customer support for your SaaS offerings.
• Make sure to market to the growing mobile customer segment.
• Constantly test and update your product delivered on a SaaS platform.

In the end, the effort to adopt the SaaS paradigm will be well worth it. You will be able to generate loyal customers, boost revenues, and position your business to achieve profitable growth targets.

The post Guest View: Why developers are struggling with SaaS appeared first on SD Times.

First things first: What comprises the Azure platform? It’s not turtles all the way down; in Azure principal program manager Scott Hanselman’s words, the underlying layer, the “infinite hard disk in the sky,” is Azure storage, where you can drill down to every virtual hard disk (VHD) image in your deployment. The next level up comprises virtual machines, which you choose, configure and manage.

On top of those VMs is a middle ground between IaaS and PaaS: Worker Roles, which are stateless cloud apps that can scale their VMs up or down. Above Worker Roles we are clearly in PaaS territory, with Web Apps, Azure Batch and HDInsight (Hadoop) for Big Data analysis. And at the top there are Web Jobs, Mobile Apps, and Media Services. These pieces are among those also available as the Azure Stack for on-premise datacenters or hybrid cloud applications.

(Related: PaaS gets a new lease on life)

“The first distinction to make as an Azure customer is, do I want to consume VMs, or do I want to consume the platform that Azure provides me?” said Esteban Garcia, Visual Studio ALM MVP and chief technologist for Nebbia Technology, and Azure company in Orlando. “Within PaaS, we typically do a Web app and SQL services. Those are pretty straightforward and easy to discover.”

It’s likely that most cloud problems you’re facing have already been solved somewhere in the Azure community, advises Corey Sanders, director of program management for Azure. “Make sure you look at the full breadth of services we offer, because they solve a lot of different problems. That’s the value of having a broad platform such as Azure,” he said.

Read on for tips from these Azure experts.

1. Saving pennies, saving dollars
The promise of the cloud is elasticity: sizing your deployment according to demand. Too often, Azure users are surprised by the multiple dimensions of pricing, from storage to transactions, support, bandwidth and more. The first step toward transparency in billing is not to “set it and forget it” but to monitor, measure and adjust frequently.

“One aspect of VMs that is not well known but that I think is super cool is the wide variety of VM sizes that we now offer,” said Sanders. “And the fact that when we stop a VM through the portal, we actually stop billing for it. The combo of those two has resulted in very few points of contention around billing.”

The Azure Billing Alert Service can create customized billing alerts for your Azure accounts, and the pricing calculator is your friend.

2. Elastic Scale for Azure SQL Database
Performance and price limits on Azure SQL Database, which offer a subset of SQL Server features, can slow you down while also burning cash.

“People use SQL Service instead of the full server,” said Garcia. “Whenever you do that, you pay for Database Throughput Units. You could be paying for 10 databases with five DTUs [database throughput units], which is a small number of DTUs to use. What you can do is start using this elastic scale, say ‘I’m going to assign 100 DTUs to 10 databases,’ and they are going to share that processing power among all the databases.” In other words, a P3 tier database costs US$4.65 per DTU per month, while the same number of DTUs scaled out rather than up on an S2 instance cost $1.50 a month.

“They are all able to use that shared pool of resources rather than being constrained,” said Garcia. “It gets along with the idea that cloud allows you to draw resources from anywhere as needed.”

Currently in preview, Elastic Scale simplifies the scaling of data tiers from just a few to thousands of database shards via .NET client libraries and Azure service templates. High-volume OLTP, multi-tenant SaaS, and continuous data collection from telemetry and Internet of Things applications are likely use cases.

3. Preview pricing
The aforementioned Elastic Scale is just one example of many new features available at preview pricing, which may be free or 50% less than the general availability pricing. Taking advantage of preview pricing lets you play with new features, stay ahead of the technical curve, save money, and possibly beat the competition by having production-ready deployments when the features go live for all customers. A list of preview services for Azure is available here.

4. The Azure Portal
Also in preview is the Azure Portal, a new dashboard for accessing IaaS and PaaS deployments.

“I was not happy with the portal at first, but now it’s growing on me,” said Hanselman in his June 2015 TechDays UK keynote. “If you double-click on the background, you can change the theme to dark. This made me so happy.”

Right-clicking on a given window pins it to the start board. Charts can be edited to show, say, CPU percentage, pricing, disk usage and more. “Don’t discount the portal quite yet; it’s fantastic,” said Hanselman.

5. Keyboard shortcuts
Every computer user knows the mouse can be deadly, in terms of ergonomics and efficiency. The best economy of movement is achieved with keyboard shortcuts. Launched with version 5.0 of the Azure Portal, the shortcut menu can be accessed by hitting shift + ?. Luckily, there aren’t too many to memorize. You’ll want to use these and more:

Hubs (left menu) shortcuts:
H – show startboard
N – open Notifications hub
A – open Active Journeys hub (a Journey is the current opened group of blades; a blade is card/tab/subpage that contains some group of tiles, e.g. website properties or analytics)
/ – open Browse/Search hub
B – open Billing hub
C – open Create/New hub

Changing focus between blades shortcuts:
J – move focus to the previous blade
K – move focus to the next blade
F – move focus to the first blade
L – move focus to the last blade

6. Azure resource manager
The complexity of managing websites, virtual machines and databases just got a little simpler with the addition of the resource manager in the new Azure Portal. Group and view resources (such as an instance of Application Insights along with a Web application and SQL database) as a single resource group. Deployment templates in Visual Studio are also aided by IntelliSense that surfaces new resource providers and template language functions to you as you write deployment templates, avoiding pesky naming errors.

“Azure resource manager has an exciting, growing community,” said Sanders. “We’re seeing in templates for the resource manager—starting about six months ago—a pretty exciting pickup in the community. We seeded GitHub with a set of these templates and put them all out fully open source. Now we have over 140 contributors and more than 200 templates available. It’s a delightful outcome to see this service that we didn’t do a huge amount of coverage on get this kind of response.
7. Scale Sets
Do you have Big Data or container-based workloads? You may want to orchestrate these complex, large-scale deployments with Scale Sets. Also in public preview, Azure Virtual Machine Scale Sets let you manage and configure virtual machines as a set of identical Windows or Linux images.

“A customer can come in and say, ‘I want VM Scale Sets in groups of 10, and I want to configure them all with a tool like Chef or Puppet,’ ” said Sanders. “The other aspect with Scale Sets that’s exciting is the deep integration with Azure Insights autoscale, which restricts cost and spending by only using the compute resources you need, responding to traffic changes.”

8. Security
A Denial of Service attack can hurt your wallet, hamstring your business and harm your customers. Like other cloud providers, Microsoft aims to share its security knowledge as well as build in basic protections. Azure Security Center’s view lets you set policies across all your subscriptions and monitor security configurations. The good news is, the days of accidentally raising DDOS flags by testing or polling your own app are over, thanks to cutting-edge threat intelligence around malformed requests and traffic sources.

9. Site extensions
Another powerful way to add custom administration features to your Web apps is with site extensions. Write them yourself or choose from the new Site Extensions Galler These live on the SCM (site control manager) site for administration and debugging that runs over SSL and is created with every Azure website. The URL for your SCM site is the hostname plus “scm”. Thus, “sdtimes.azurewebsites.net” would have a corresponding SCM site at “sdtimes.scm.azurewebsites.net”.

10. PowerShell cmdlets
Whether you want to clean up a deployment where you have some extra VHDs and VMs lying around, provision VMs, set up cross-premises networks, or other production tasks, you’ll enjoy the Unix-like scripting power of PowerShell and the new PowerShell cmdlets. As of the November 2015 update of the Azure SDK 2.8 for Visual Studio 2013 and Visual Studio 2015, the PowerShell script for deploying Azure Resource Manager templates now works with PowerShell cmdlets. Find scripting solutions already crafted for you in the PowerShell Gallery.

11. Service fabric
Riding the microservices revolution, Service Fabric is Azure’s platform for assembling cloud applications from a large collection of services. “Service fabric offers a platform that runs on Azure but also on-premises,” said Sanders. “This is a platform for deploying, managing and maintaining microservices. Discovery is handled for you, and it supports stateful and stateless microservices.”

12. Docker
The explosion of ways you can tinker with cloud resources, from remote desktops and SSH to portal shortcuts, has only just begun. According to Hanselman, the microservices revolution means there will soon be even more options to choose from.

“Simply stated, if I’ve got a tiny little 10MB PHP app sitting inside of a 5GB VHD, that’s a lot of VHD, a lot of virtual machine for a small Web application,” he said. “Does it really need that weight? That much security and isolation? It just needs to be in a container, and it needs to be deployable in a reliable way. Docker will provide that.”

“The excitement around Docker is very real. It makes it incredibly easy to deploy in ways that have never been possible,” said Sanders, who notes new integration of Docker support into Visual Studio and the Azure marketplace.

“My biggest tip and trick with Docker containers is just to deploy one. If you’ve never done anything with Docker, there’s a way to quickly deploy with a fully packaged VM and Docker in the Azure marketplace. No bringing down of the Docker engine, no pulling down the hub.”

13. Azure DevTest Labs
How do you avoid using up all your MSDN credit while testing on Azure? The preview of Azure DevTest Labs lets you spin up Windows and Linux environments to deploy and test applications while avoiding cost overruns.

“With Azure Dev/Test labs, the idea is that a lot of times developers have to wait for someone to spin up labs for them,” said Nebbia’s Garcia. “This allows you to spin up environments much quicker. You can choose for it to run a maximum eight hours, and after that it gets shut down. It’s a quick way to provision environments but avoid the problem of leaving it up and running. You can push a button and have whole sandbox.”

14. Application Insights
“Application Insights allows you to dig down and find the root cause of any application issues and understand how people are using the application,” said Garcia. “I’ve been using it for a year and half…as a Microsoft MVP.” For example, he uses it for availability testing from different geographic locations, either as a static test that checks a single page, or as a test of dynamic application flow.

15. Kudu, CloudBerry and Sendy
The Kudu open-source project is a useful troubleshooting tool and client-side process explorer for capturing memory dumps or looking at deployment. It’s also a site extension and welcomes community participation.

Another useful freeware tool is CloudBerry Explorer for Azure Blob Storage, which offers a file manager-style user interface to Azure Blob Storage.

If you’re already mucking around in the cloud, you may have e-mail update needs that can be met by Sendy or similar tools. Sendy was designed to work with Amazon Simple Email Service, but can be adapted for Azure as well. The cost savings versus a hosted e-mail solution such as MailChimp can be enormous.

16. Remote debugging
In its September 2015 white paper, “Practical Guide to Platform-as-a-Service Version 1.0,” the Cloud Standards Customer Council notes that no PaaS worth its salt should be without remote debug capabilities. “Application developers should have access to tools that enable them to control activities in the PaaS—for example, uploading (‘pushing’) application code, binding services to applications, controlling application configuration, starting and stopping application instances,” it said.

“Such capabilities should be provided in a way that fits well with the other tools used by the developer—command-line tools, graphical tools, embedded components for development environments. Ideally these tools should work via an API that is exposed by the PaaS system—cloud service customers should look for these APIs and assure themselves that the API can be used by a variety of custom tooling code.”

Remote debugging with Visual Studio fits the bill: Developers interact with cloud applications as if they were on-premise. Best used with Visual Studio 2013, remote debugging lets you manipulate memory, set breakpoints, and step through code—with the caveat that breaking a running process could break your live website. Save this one for pre-production sites.

17. Performance testing
Another public preview that is currently free to use, performance testing, allows you to generate thousands of virtual users from around the world and test your application against the load.

“We started using performance testing in the past six months,” said Garcia. “If you spin up a Web application, you’re able to do a performance test right from Azure, right in the cloud. Before, it was more on the Visual Studio side. So I can see what it looks like if 1,000 people hit my app at once. It’s very useful in knowing how to scale the application: We can have fewer servers, but make them stronger by adding this performance testing feature right within the Azure portal when you first launch an application.”

18. Easy ROI: Lift and shift
Want an instant return on your cloud investment? Eliminate idle servers that only handle periodic loads. “This is something people forget about when they’re thinking about the cloud,” said Hanselman. “Azure storage, that’s an infinite disk that’s out there. You probably have a machine sitting under your desk and it’s got a VM, running maybe an expense reporting system. It’s something that you need to lift and shift into the cloud. There are migration tools that can help you. Literally, it’s Hyper-V in the cloud, but that’s only the most basic way of using Azure. So, Step 0: Lift and shift. Then start thinking about other ways to exploit stuff.”

19. Developer services marketplace
Before you reinvent speech recognition, check the Developer Services Marketplace for free and paid ways to extend functionality, turbo-charge development, and manage cloud deployments with certified Azure tools such as io for event-driven computing, or face APIs from Project Oxford.

20. Ride the IoT wave
No set of tips would be complete without instructions on how to program the proverbial light bulb. These days, Internet of Things projects are everywhere. Hanselman, a type 1 diabetic, movingly demonstrated in a November 2015 keynote video how he tracks his blood sugar and insulin pump in the cloud with Azure technologies.

Microsoft Azure IoT Hub offers SDKs, management and security solutions to harness a plethora of IoT devices, and once the data is collected, there are new machine-learning tools available to process data stored in HD Insight, Microsoft’s version of the Hadoop Big Data store. Have fun!

Start your engines
Use these 20 tips as a checklist for leveraging the vast Azure platform. The more you understand Azure, the more you see where it’s headed: “We’re seeing a blurring of IaaS vs. PaaS and starting to just see a compute platform,” said Sanders.

Redmond has clearly learned to embrace today’s polyglot cloud, and you can use that flexibility to your advantage. “Azure is not only a Windows server; they have Ubuntu, they have Linux servers, they have a new agreement with Red Hat, you can spin up an Oracle database… There’s so many different non-Microsoft technologies to choose from,” said Garcia.

The post 20 ways to build up your Azure deployment appeared first on SD Times.

Built with DreamFactory’s open-source REST API MBaaS and hosted on Verizon’s cloud, the DreamFactory-Verizon Partner Solutions Portal gives developers a cloud-first development environment for desktop, mobile and IoT app creation. The portal provides enterprise developers with CloudSpace sandboxes to build application workloads, offering capabilities such as DreamFactory’s REST back ends, LAMP (Linux, Apache, MySQL and PHP/Python/Perl) stack, MongoDB integration, and external APIs and services offered by Verizon Cloud.

The portal is designed for DevOps and Dev/Test methodologies as well. Dev/Test capabilities have been built into DreamFactory sandboxes for Verizon Cloud with prepackaged REST APIs, along with Ops deployment in CloudSpace and reporting/tracking metrics.

(Related: DreamFactory wants to stop the madness of building your own APIs)

Bill Appleton, the CEO of DreamFactory, said a partnership with Verizon made sense in what he called a “Reese’s peanut butter moment,” combining DreamFactory’s REST middleware with Verizon’s public and private cloud deployments to create an end-to-end DevOps and IaaS platform for developers.

“Developers are increasingly in the driver’s seat when it comes to making choices on which clouds and platforms to use,” he said. “If you can win developers onto your platform, the easier it is to grow your business. Combining DreamFactory’s REST API back end with Verizon’s enterprise-grade cloud services in a free development portal enables both parties to attract more developers and work to grow them as customers.”

According to Appleton, DreamFactory will soon release a commercial enterprise offering to manage multi-tenant production instances, and the partnership with Verizon gives the REST API platform a foothold in the enterprise cloud market. The Verizon partnership and development portal will serve as DreamFactory’s gateway to what Appleton sees as a shift away from pure MBaaS and toward more flexible hybrid MBaaS/IaaS cloud-based mobile and IoT development.

“In the longer-term, we see this as the fastest, most effective way for enterprises to securely mobilize data for the new generation of mobile and cloud-based productivity applications for their employees, partners and customers,” said Appleton. “Securely mobilizing data has been a key challenge for companies looking to develop these new mobile applications. DreamFactory running on Verizon’s secure cloud platform provides a huge productivity boost for the development and operation of these applications, speeding time-to-market, and driving innovation.”

The post DreamFactory and Verizon partner on enterprise cloud development portal appeared first on SD Times.

Yes, the language repository was posted on April Fool’s Day by mobile application development company Hummingbird Technologies. Yes, the language’s structure and “simplified” syntax consist entirely of the word “Hodor” in tandem with various symbols and punctuations.

Yet despite all that, within the software development, open-source and SD Times readership communities, there exists a deep, passionate following of “Game of Thrones” and “A Song of Ice and Fire.” And they want—nay, need—a Hodor programming language.

Technically written in JavaScript and run on Node.js, Hodor is a language for all. Hodor does not discriminate. Hodor doesn’t care whether you’re a coder from The North or from Dorne, from House Arryn, Lannister, Martell or Tyrell. Hodor knows no allegiances, not to programmers praying in reverence to The Seven, the Old Gods, the Drowned God or Red R’hollor the Lord of Light. Hodor simply is Hodor: a blissfully simple programming language for all of Westeros, Essos and the modern world.

“A collaborative effort by DARPA, IEEE and Hummingbird Technologies since the early 1980s, Hodor is genuinely the perfect programming language,” according to its wiki. “Implemented by Governments, NGOs, the Singularity Institute, and billions of companies worldwide, Hodor answers all questions emphatically and succinctly: ‘Hodor.’ ”

Hummingbird Technologies has not officially agreed to the next point, but for the coder who programs the most innovative application written in Hodor, he or she shall be awarded the prestigious…

The robust JavaScript implementation can be used to develop Web and mobile applications, or to create servers. It even comes with a js2hd compilation tool to convert drab old JavaScript files to the splendiferous, streamlined diction of Hodor.

Hodor’s language features are as follows:

1. Hodor
2. Hodor hodor hodor hodor
3. Hodor hodor hodor
4. Hodor HODOR hodor
5. Hodor!

Hodor on, o’ courageous and valiant coders. Hodor on.

Top 5 projects trending on GitHub
#1: Facebook’s React Native framework, was the subject of last week’s GitHub Project of the Week.

#2: Fresco, another Facebook open-source project announced at F8, is detailed in SD Times coverage here.

#3: Free-for-dev is a collection of SaaS, PaaS, IaaS and other free developer service offerings on the Web.

#4: Airbnb’s JavaScript Style Guide covers JavaScript from beginning to end in the context of how the popular international crowdsourced lodging site deploys the language.

#5: Termui is a terminal dashboard written in Go, offering cross-platform functionality, easy compilation, and full customization. Not much more to it than that.

The post SD Times GitHub Project of the Week: HODOR appeared first on SD Times.

The “3rd Platform” IDC is referring to, which it coined back in 2007, is the concept of the next generation of software founded on cloud computing, Big Data, mobile devices and social media. (Gartner’s “Nexus of Forces” is a similar concept.) The growth of the “3rd Platform” in all phases of the information and communications technology (ICT) market is a major theme of the forecast, but 10 of IDC’s predictions in particular stand out, spanning growth in mobile, cloud, Big Data, the Internet of Things and more.

1. New technology will take over the market: According to IDC’s predictions, worldwide ICT spending will grow 3.8% in 2015 to more than US$3.8 trillion. Nearly all of this spending growth will be focused on 3rd Platform technologies, while 2nd Platform technologies (PCs) will continue to see declining spending.

2. Wireless data growth: IDC projected that wireless data will balloon to 13% of telecommunications spending, accounting for $536 billion. The company predicted this will lead carriers to develop platform- and API-based services to attract developers.

3. Phablets will rise, wearables will underwhelm: Mobile device and app sales will reach $484 billion in 2015, according to IDC, accounting for 40% of IT spending growth. This is a slower, steadier rate of mobile growth than in past years, and IDC projected it will be driven largely by phablets and enterprise mobile app development. The firm added that wearables will see an explosion in innovation that will not translate to market value quite yet. At the same time, IDC believed Chinese vendors (such as Xiaomi and OnePlus) will capture 15% or more of worldwide market share.

4. Cloud services: IDC predicted PaaS, SaaS and IaaS services will remain a hotbed of activity in 2015, with $118 billion in spending. IaaS adoption is projected for the highest growth (36%) as new competitors take on AWS. According to IDC senior vice president and chief analyst Frank Gens, IDC foresaw “new partnerships forming among ‘strange bedfellows,’ such as Facebook with Microsoft and/or IBM or Amazon partnering with HP, to expand market opportunities.”

5. Big Data and analytics: IDC predicted Big Data spending will grow to $125 billion in 2015, with video, audio and image analytics taking on more importance. The firm also projects that Data-as-a-Service will drive the forging of new Big Data supply chains focused on commercial and open data sets.

6. The Internet of Things: The predictions identified IoT as one of the most important factors for growth of the 3rd Platform in 2015. IDC predicted the invention of more intelligent and connected “things,” with a third of IoT spending focused on intelligent embedded devices and predictive maintenance emerging as an important IoT category.

7. Cloud services will become the new data center: IDC said that data centers are undergoing a fundamental transformation in the 3rd Platform era, with compute and storage capacity moving to cloud, mobile and Big Data-optimized hyperscaled data centers operated by cloud service providers. The report forecasted that this shift will spark a burst of “cloud-first” hardware innovations that consolidate server, storage, software and networking, leading to two or three major mergers or acquisitions.

8. The 3rd Platform will become industry-specialized: IDC predicted that 3rd platform technologies will specialize into industry-specific enterprise platforms, doubling their number in 2015. The company mentioned examples such as alternative payment networks in financial services; expansion of IoT technologies into city safety, public works and transportation systems; and the expansion of location-based services in the retail industry.

9. Security and 3D printing: IDC sees “3rd Platform-optimized” security solutions for the cloud, mobile and Big Data as a growth sector to secure these technologies, with mechanisms including biometrics on mobile devices and encryption in the cloud, as well as threat intelligence emerging as an essential Data-as-a-Service category of enterprise-specific threat information. IDC also projected that 3D printing will gain traction in conventional document printing companies, which will come to fruition market-wise in 2016.

10. Massive growth in China: China will experience skyrocketing influence on the global ICT market in 2015, according to IDC, with spending accounting for 43% of all industry growth, a third of all smartphone purchases on the worldwide smartphone market, and about one third of all online shoppers. IDC predicted China’s cloud and e-commerce leaders such as Alibaba, Tencent and Baidu will become global market powers.

The full “IDC Worldwide Predictions 2015: Accelerating Innovation on the 3rd Platform” report can be found here, and more information is available in IDC chief analyst Gens’ webcast.

The post IDC’s Top 10 technology predictions for 2015 appeared first on SD Times.

The barriers to entry are now so low that the value of being savvy with where and how you build your presence is no longer up for debate—it’s a necessity for any type of organization. Today, there are more than 2 million mobile applications, and the majority of organizations in the U.S. own a domain and host a website.

Whether building an online or mobile application for yourself or for a business, you need to ask yourself: What are the core offerings from the hosting provider that will give me the most value out of my digital presence?

It’s no secret that cloud hosting services aren’t hurting for company. With many players vying for the largest market share, smaller companies are growing quickly and will live or die by how they differentiate themselves. Some of these smaller players have turned away from their bread-and-butter foundation (raw infrastructure) and have instead turned their focus to other offerings such as managed services.

While this approach certainly makes sense for businesses in certain situations, for developers the most important thing is that their cloud provider focuses on providing the best infrastructure experience possible. Here’s why:

Infrastructure is the common building block of any online business. All digital businesses begin by writing their code and applications for the market on a personal computer or laptop. But ultimately, when they are ready to push it live, they need to set it on a reliable foundation. Infrastructure is the life force of the data that every Web or mobile application relies on. It needs to be strong enough to hold the pieces together when an application goes to market, and it needs to have the ability to expand at the same rate as its use.

Developers staking a claim online don’t need anyone to hold their hands. Developers are equipped to navigate their path and need little support, as long as the foundation they’re building on is solid and can grow alongside the application.

This, combined with the myriad tools now available, make it so developers have the resources they need to manage the use of infrastructure on their own. The infrastructure experience should be designed to be a road map. The more convenient the map is to follow, the less time is wasted with distractions along the way and the more time is spent building something of value.

Instead of investing in contingency plans, cloud providers working with developers should focus on providing the most user-friendly, scalable infrastructure possible. There’s no need for call centers if the path is clear.

Layers of management slow down provisioning times and get in the way. Of course, cloud-management services need to provide some level of support. But businesses depend on how their sites function and how quickly they can scale up or down in order to adjust for market demand and traffic waves. What developers really need are things like quick provisioning times and straightforward pricing.

IaaS will continue to evolve. We’ve really only begun to develop sophisticated infrastructure in the past few years. It will continue to change and grow and develop, and as it does, the hope is that it will become both more robust and more straightforward.

As more players in this space realize the importance of focusing on the developer, cloud-management services would do well to focus their efforts on creating what developers actually want: a feature-rich cloud that remains simple. When we provide developers with what they want, we don’t need to get in the way.

The post Guest View: The ideal IaaS gets out of the developer’s way appeared first on SD Times.

Working with people that aren’t part of the choir, however, shows that cloud is still one of the most misunderstood phrases in the IT industry today. In some ways this is no surprise, too, because unless you’re a direct customer working in each layer of the stack, you’re probably finding it hard to conceptualize how the layers fit together beyond the popular cloud pyramid diagram. That hasn’t stopped people from spending on cloud-based services as a recent report from Juniper estimates that global spending on cloud technology will go up 20% from last year to US$174.2 billion, and will reach $235.1 billion by 2017.

More recently, the confusion over cloud-computing layers has manifested itself into some pretty heated, wildly misleading declarations, particularly at the PaaS to IaaS “boundary” (which is double-quoted because it’s an artificial dependency). Over the past year or so, we’ve seen more than a few articles and reports that predict the extinction of PaaS, as IaaS vendors go upstack to offer more and more PaaS-like features. Although I respect these analysts and IaaS vendors (some of whom are well renowned in the cloud-computing world) for speaking their minds, they can’t be farther from reality.

SaaS vs. PaaS vs. IaaS
Let’s start with some quick-and-dirty definitions of each layer in the technology stack. SaaS is easy for consumers to understand, but the distinction between PaaS and IaaS is much harder for non-developers to distinguish.

Software-as-a-Service is the easiest to understand because it’s widely available to consumers and business users, and has been made a common word by companies like Salesforce.com. SaaS refers to a Web-based piece of software that customers can start using right away, and it is a modernization of traditional business applications. Typically, a SaaS application is an online app that a user logs into and uses. The vendor releases software updates and patches to this central app, and all data is stored in a managed cloud. Gmail and Mint are other examples of SaaS applications.

Platform-as-a-Service is probably hardest of the three for the uninitiated to comprehend, because these services are built for developers. PaaS is a computing platform upon which devs can build their applications without having to worry about configuring any software or infrastructure underneath, or without having to worry about how the application they built needing to be written to run in a cloud-like environment.

(Related: How to pick your best PaaS)

As a result, developers can spend 100% of their time on their core competency of developing applications, and never again worry about how an application needs to be built for a cloud world, how it needs to be run, or the details related to managing it. PaaS is a modern version of the application server (think JBoss). For example, my company, Apprenda, provides a PaaS software package for enterprise developers.

Infrastructure-as-a-Service is at the bottom of the cloud stack. IaaS refers to all the servers, storage, networking and operating systems needed to run your applications at scale. IaaS simply provides this infrastructure on demand, eliminating the need for your own employees to maintain everything. IaaS is the new hardware. IaaS is great for those who need—but can’t afford—high-performance computing power with a few clicks of a button.

Why PaaS and IaaS should stay separate
I can see why analysts think PaaS will be subsumed by IaaS in the near future. Articles by David Linthicum and Alex Freedland of Mirantis describe PaaS as a natural extension of IaaS, only providing the ability to add some extra functions on top of existing infrastructure such that developers can easily work in one environment, and their colleagues in IT can control the infrastructure.

In reality, PaaS is a feature of IaaS as much as WebSphere or JBoss is a feature of Linux OS—not at all. PaaS specializes in a platform that supports the writing of code, which cannot easily be classified as a commodity.

If you consume PaaS as an extension of IaaS, your end result will be blunt stack lock-in, which no one wants. Blunt stack lock-in means you sacrifice infrastructure independence. Why? Because features will be tied to a specific IaaS layer.

For the sake of playing devil’s advocate, aren’t there key benefits to a tightly integrated PaaS and IaaS? Yes, it can lead to better, faster deployment of applications. But this is why PaaS vendors like Apprenda have achieved this through secondary integration, not as an explicit dependency, which leads down the lock-in path.

Don’t pass on PaaS
By keeping your PaaS and IaaS vendors separate, customers have freedom to optimize layers independently and not be shackled by a “downstack dictator.” If innovation is what you’re after, you’re far better off limiting your infrastructure than limiting your development environment.

PaaS offers many unique capabilities that should remain independent of IaaS, and Fortune 500 companies like AmerisourceBergen, JPMorgan Chase and McKesson, are realizing these benefits today. This is why PaaS will be here for a long time, and over the next few years we will see large enterprises driving PaaS adoption and hybrid cloud usage: Two trends that, in my opinion, go hand-in-hand.

The post The case for separation of church and state regarding PaaS and IaaS appeared first on SD Times.