Interested in how DevOps, IT Modernization and Agile practices can positively impact customer experience?
This week, we’re starting a new project with a client who is laser-focused on improving the performance of its website. This ecommerce company knows its customers won’t stand for slow load times. And, a poor-performing site can damage its search rankings, risks it doesn’t want to take. Small and large businesses alike are increasingly migrating their mission-critical business functions to the cloud. They are increasingly linking their success and business agility with their ability to maintain high-performance websites and services. While cloud service providers strive to maintain 99.9% uptime rates, risk still exists. They are in regard to service providers and related to how a business’ cloud infrastructure is architected and maintained. Most of us use cloud or web-based (SaaS) ecommerce, games, business applications, data storage, and social or information services on a near basis. And, while many of us rely on these applications or cloud services to run our own business, we won’t stand for poor performance. Read more here: 3 strategies to fail fast, fail cheap & get agile Why startups are choosing AWS 4 tips to better decisions for cloud benefits vs cloud computing disadvantages Businesses that rely on large, time-sensitive or high volumes of web-based transaction (such as brokerages) are even more significantly impacted by downtime. Other organizations, such as those aiming to achieve continuous integration and continuous deployment practices using web-based development, can experience significant disruption when web services are down. What’s more, the ecosystem of businesses dependant on website or web service uptime is expanding. Increasingly, new business models are relying on performance to maintain their revenue streams and keep their sites highly ranked on search engines. Four new markets that can’t compromise on uptime include: Sharing Marketplaces: Uber, Airbnb, and, now, B2B services are experiencing rapid growth.
It’s making both business and trade headlines. You are reading all about it every day. What is it? The constant and steadily increasing buzz from organizations of all sizes about the business advantages and possibilities of adopting a cloud infrastructure. No doubt, the cloud is increasingly seen, not just as a way to cut the costs of hardware, but as a way to drive business growth. However, there’s still an element of fear, uncertainty and doubt about migrating a business’ infrastructure to the cloud. It tends to haunt business leaders and IT alike. Even seasoned IT pros are surprised and overwhelmed by cloud providers, their extensive lists of services and features, and cloud architecture options. On the other hand, there may be many questions during a cloud migration that are left unanswered. IT teams are often left to fend for themselves. This latter point by itself is a good reason to delay decisions and engage with a reputable cloud consulting organization. Not doing so, can lead to disaster. A few common questions you might have include: What are cloud security best practices? Will my organization be able to meet industry standards of cloud compliance? What is the expected extent of vendor lock-in? These are questions that should be considered after you commit to a cloud strategy and before you choose a cloud vendor. But first, it’s smart to really understand whether the cloud is right for you. Will your business benefit from this transformational trend? Cloud migration can be the first step toward changing your organizational culture. It requires a basic understanding of how the cloud affects your business and why you should, from a business perspective, want to move to the cloud.
The Greek philosopher Heraclitus once said: “The only thing that is constant is change.” No truer words have ever been spoken, And, your business is no exception here. New technologies are constantly replacing older ones. We are in the era of technology advancements. And, this defines the existence of businesses. Your business is under constant pressure by the digital natives or Gen-Z. Business requirements are no longer predictable. Making your customer happy is not a cakewalk. On the other hand, your list of competitors outruns your list of customers. According to the industry analyst firm Forrester Research: Every industry is subject to disruption. Only a true agile business is equipped to respond. Indeed. Companies that existed a few years back have become dust because of their lack of business agility. While you may have the best team that’s booming with ideas, a misalignment with the actual trend or customer requirement is fatal to your business. So, what’s this business agility that everyone keeps talking about? And, how can you make sure your business is capable of handling this fast-growing necessity? Business agility as defined by Forrester is: The quality that allows an enterprise to embrace market and operational changes as a matter of routine. In other words, the mantra to adhere to business agility is to follow a “fail fast, fail cheap” strategy. Similar to any DevOps approach, business agility is a process or culture change. While there are a lot of technologies and tools to help optimize or improve business agility, they can’t be THE go-to solution. Tools and technologies only complement the strategies you put in place to adapt to changes.
At the New York Times, one intern recently learned a painful lesson: It can take a long time to set up a development environment. In this case, about two weeks. While set-up time varies for a number of reasons, can your business afford even half of those cycles every time your developer needs to test something new? Sure, it’s the New York Times. This intern is probably dewy-eyed and ambitious. Wait, isn’t that good? Don’t you wish you could capture that passionate spirit of creativity and innovation for your startup? Are innovation and efficiency juxtaposed in the world of software development? Or, can you have your cake and eat it, too? That’s just what Djed Studios was aiming to do when the consumer Internet company created push-button development environments using technologies, including Docker. Djed Studios’ stack is in Python, a skill set lacking in most developers. A push-button reset and automated code deployment process helps developers test the impact of new code, create a streamlined web development workflow, and foster a self-service culture that reduces the workload for the infrastructure team. Eat your heart out New York Times. Read this turnaround story today about how Djed Studios leveraged Docker to improve web developer productivity. Right now, app development, specifically for mobile, is one of the hottest jobs in IT, according to CompTIA and as reported in this FierceCIO article. These developers must test their app on any number of constantly changing platforms. That means a whole lot of test and Q&A cycles. In order to create efficient, high quality (i.e., non-buggy) apps, mobile app developers must test in their local development environment as much as possible.
Against the backdrop of repeated and well-publicized data and privacy breaches, businesses are continuing to evaluate cloud benefits against cloud computing disadvantages. The tradeoffs are difficult: streamlined, cost effective infrastructure vs. loss of control. But cloud infrastructure is not sorcery or black magic. While you can expect a whole new level of automation features out of the cloud, expecting the cloud to automatically fix issues for you is unreasonable. The cloud has come a long way over the past few decades. What was once a dream is now achievable for most businesses. Media Darlings: Cloud Computing Fails While the tremendous advantages and features of the cloud are now largely understood, the media frequently focuses on the setbacks of a virtual infrastructure when it’s not properly set up. It was roughly a month ago when we learned that Code Spaces shut down because its AWS account was hacked by the bad guys. You can read the full story here. While we are just recovering from the chills of this incident and others, the very recent security breach of celebrities’ iCloud accounts is in the headlines. On one end, Apple denies that the hack is linked to a security breach of iCloud. While on the other end, it has stirred up serious debates in the community questioning the reliability of the cloud. Two Sides of the Cloud We certainly can praise the advantages of the cloud and the features it has to offer. By using the cloud the right way, you can end up revolutionizing your industry. It’s enticing at various levels by simply addressing issues at technical and business perspectives.
Amazon Web Services (AWS) EC2 offers a variety of instance families and types, each serving specific use cases. The instance families include: General Purpose: T2, M3 Computer Optimized: C3 Memory Optimized: R3 Storage Optimized: I2 GPU: G2 In this post, let’s pick the general purpose M3 family. It’s a balance of compute, storage and network services. Let’s analyze it in depth. M3 is the new generation of the the ‘m’ family, while m1 is the previous generation. So, first things first. Let’s refresh some M3 basics: Best suited for: Data processing, caching and running backend servers. Features: Processor: Intel Xeon E5-2670 (Sandy Bridge or Ivy Bridge) Storage: SSD Let’s analyze the following with respect to m3: CPU Performance To understand CPU performance of m3 instances, we used CoreMark, an industry standard benchmarking suite. Read the experiment setup here: CoreMark setup. The average CoreMark score for the m3 instances are tabulated as follows: Instance Type CoreMark Score Std Dev Std Dev as % CoreMark /$ per hour m3.medium 6879.731 30.062 0.4369 60882.577 m3.large 19539.593 101.749 0.5207 86842.639 m3.xlarge 36699.549 75.971 0.2070 81554.553 m3.2xlarge 73522.869 579.416 0.788 81692.077 The following graph shows the CoreMark score for the m3 instances: It’s pretty clear that the CoreMark score per dollar is off the charts for m3.large. This means that it offers the best performance per dollar. Disk Bandwidth FIO, a tool for benchmarking and stress testing IO, was used to analyze the disk bandwidth of m3 instances. Read the experiment setup here: FIO setup. Four IO operations were benchmarked using the tool.
In August 2013, Wired wrote a controversial article that focused on why some startups were getting away from the cloud and returning to on-prem solutions … all because of costs. The end of cloud-based startup infrastructure, however, wasn’t near. In fact, one of the challenges cited in the Wired article — costs — got easier to deal with this year with falling prices. Read our blog on what the AWS price cut means to you. We continue to see a steady stream of startups moving to Amazon Web Services [AWS] from other providers. No doubt, AWS is difficult to beat for costs, stability, and its robust tool set. Recently, one stealth startup was creating an app it hoped would become very popular and eventually have millions of users. It would have been cost prohibitive to setup a static environment during day one that would accommodate that many users. Company leaders knew they needed something that could scale. After evaluating a few other cloud platforms, they decided AWS would be the best fit. However, they were still missing the cloud architecture knowledge they needed. We worked with them to ensure their setup was optimized and ready to be tested for success. The co-founder of the stealth startup said: “Without Flux7’s help, we wouldn’t have been able to set up such a well-structured account, and we surely would have encountered issues upgrading in the future.” Similarly, another startup, poised on the brink of a major marketing campaign, realized that before it pressed the launch button, it needed to have some solid infrastructure in place. And Heroku just wasn’t it. You can read about this startup’s challenges in detail here. Two’s Company … Three’s a Trend? Yet, another startup worked with us to move its infrastructure from another vendor to AWS.
We love Amazon. We also recognize, along with industry research groups like Gartner, that Amazon is the cloud leader in regard to its scalability, security, flexibility and costs. But, there is a downside to Amazon. And many organizations are discovering it too late in the game and exposing themselves unnecessarily to business risks. Amazon is by far the market leader with more than five times the computing capacity of the other 14 competing providers. The Amazon platform breadth and depth means it can support sophisticated, scalable applications. It’s globally available and can be integrated effectively with on-prem data centers to provide hybrid cloud support. What’s more, Amazon continually upgrades its infrastructure. It provides new services and improves its pricing models, which, when managed properly, can significantly reduce costs. So, what is Amazon’s weakness? We’ve done a lot of analysis. What it comes down to is a communication problem and a dramatic difference in perspectives that can lead to cascading failures. There is a reason that Network World recently listed AWS Certified Solutions Architect as one of the hottest jobs and highest-paid certifications in the industry (see slide 3 in the NWW presentation). And cloud architecture knowledge is in high demand. There is extensive value in Amazon knowledge both in the efficiencies it can bring and in the reduction of risk. Do You Speak Amazon? To start, Amazon doesn’t speak the language of developers. Amazon was built for operations people. As such, there is a vast terminology gap that at best leads to lost productivity from a need to perform endless Google searches for definitions. But, it’s not just developers who are at a loss over Amazon set ups. Operations folks might have a leg up, but they are often missing out on key concepts of cloud computing.
The buzz about Docker has risen to a roar during the past year. Irrespective of the size of the organization, it has attracted a lot of interest. Not convinced? Docker 1.0 is now used in production by three large financial organizations, including one that treats security as its highest priority. The welcome mat for Docker has not been restricted to just the open source community. It has pretty much impressed every rung of the organizational hierarchy. Use Docker … attract investors! Last week, Tutum raised $2.65-million for its cloud infrastructure services using Docker containers. Add to that list CoreOS, Quay, Orchard … and the list doesn’t stop there. Like others, does your startup relate to any of these issues?: Your code pipeline management is weak and ineffective. It’s a constant bottleneck. There’s a lot of inconsistency across environments. Your local dev environment and production environment aren’t similar. This leads to bugs during production. Resource provisioning takes ages. Your bills are through the roof as you deploy new servers now and then based on increasing needs. So, what is Docker and where can it help you? In simple terms, Docker: Makes Linux OS isolation tools easier to use. Looks like a virtual machine. Supports application development at a whole new level. The Holy Grail of virtualization is the use of abstraction layers. By adding an abstraction layer, this allows for the needed communication while isolating sources of contention. This is depicted in the image below. The idea is to: Allow communication Prevent conflict over resources Create an abstraction layer continuum However, to achieve this, there are a lot of costs and needed resources, including overhead, increased latency and time. This is when you reach out to Docker for a solution. Docker provides isolation features and enacts a VM-like environment.
In our last post about wikibench benchmarking, we discussed the benchmarking methodology and gave a detailed analysis of the results for the m3.xlarge instance. In this post, we will compare the wikibench performance on all ‘m3’ instance types. We ran experiments on all the ‘m3’ machines using the same methodology as in the m3.xlarge instance. We compared the performance in each of the following categories: Missed Deadlines Request Timeouts Average Response Times CPU Usage Memory Usage Missed Deadlines Due to the delayed response time from the server for lower sampling rates, we see that wikibench is not able to replay all the requests in the tracefile. This results in missed deadlines. Here is a table of missed deadlines for each of the instance types. Table 1: Number of Missed Deadlines Sampling Factor Total Number of Requests m3.medium m3.large m3.xlarge m3.2xlarge 980 46239 31726 10668 952 628 981 43847 28367 5774 267 62 982 41589 25683 4524 191 35 983 39319 23366 3277 108 15 984 36901 20888 2240 74 15 985 34725 18939 1672 43 8 986 32370 16848 1069 28 4 987 30069 14681 659 19 1 988 27720 12028 391 6 3 989 25497 9966 195 3 0 990 23271 8146 98 0 0 991 20767 5463 28 1 0 992 18340 3408 4 0 0 993 15862 1668 0 0 0 994 13520 670 0 0 0 995 11326 230 0 0 0 996 9053 46 0 0 0 997 6797 2 0 0 0 998 4505 1 0 0 0 999 2193 0 0 0 0 Request Timeouts We configured the benchmark to have request timeouts at 10000 milliseconds. When the requests rates are high, we did see quite a few request timeouts.
Having a DevOps approach for application or product development is like H2O to your organization. It’s a basic need to live long and prosper. Yes, DevOps is like H2O: High-quality bug-free apps and products Heights of innovation One-click deployments And a lot more to be honest! Whether you’re a dev or ops guy, chances are you are constantly being educated about a DevOps approach or methodology. You might even be aware of the myths and facts of DevOps and how a DevOps infrastructure affects your customers. The questions to ask are these: How do you know you aren’t using DevOps? And, isn’t it time you think of an investment in DevOps? Because DevOps is a culture –a process, there’s no definite checklist that you can use to understand your DevOps stand. The current state-of-the-art DevOps approach does not provide a ready-to-use framework or manifesto. However, there are a lot of best practices and indicators that can help you understand where you stand when it comes to DevOps. We’ve put together a list of five signs, or indicators, (it’s not THE list, as there’s no specific checklist) that, when relatable, implies you’re missing out from DevOps at the very core. In that case, you need to call for a team meeting to start thinking DevOps. Sign #1: Your team has the best developers, but an application deployment takes three times longer than it’s supposed to take. This clearly indicates that your feedback loop is at stake. Whether it’s a local feedback loop, wherein the developer does local testing before the code is sent to QA, or a feedback that a developer receives from QA, it’s significant to send feedback to the developer as quickly as possible. Why? Context switching.
Recently, Foote Partners, an IT benchmark research and advisory firm, released data on the most in-demand skills and certifications in the IT market. We weren’t surprised to see cloud architecture experts topping the list, or to be specific, architecture and cloud skills. An article about the survey is here, from FierceCIO: “Architecture and cloud skills top IT employer ‘most wanted’ lists.” In our cloud consulting work at Flux7, we see this to be a real need in enterprises, mid-sized organizations, and particularly in startups. Naturally, as the cost of skilled labor in an area such as cloud architecture increases, cash-strapped startups start to feel the pinch. But, these startups feel the effect of poor choices from the lack of cloud architecture expertise just as much, if not more, than enterprises. We’ve recently seen several cases where startups have launched environments in Heroku because of it’s ease of use, but then found the need to migrate from Heroku to AWS in order to meet compliance or scalability requirements. We’ve also seen a number of situations where a startup or mid-sized organization was set up in AWS, but needed to get this environment production-ready. Often, a bit of planning and work upfront could have saved these startups time and resources to execute a cloud migration. And, it’s rare to start a cloud migration until the pressing need is there: such as that critical demo for an investor or customer; or even an anticipated sudden increase in demand from a marketing campaign. Because startups need access to cloud architecture experts, too, we’ve created a new service: Cloud Attune. Cloud Attune helps incubators, VC firms and bootstrapped startups.