Welcome!

SDN Journal Authors: Yeshim Deniz, Elizabeth White, Liz McMillan, Pat Romanski, TJ Randall

Related Topics: @CloudExpo, Containers Expo Blog, Agile Computing, Cloud Security, @DXWorldExpo, SDN Journal

@CloudExpo: Article

Achieving the Full Promise of the Cloud

An exploration of the business value of application defined networking

As the cloud market explodes, a distinct chasm has become apparent between the operation of infrastructure and applications. Sharing, isolation and load balancing issues in the network, combined with high density virtualization in compute and storage resources, can adversely impact the performance of applications running across the network, frustrating application developers and end users alike.

Enter application defined networking (ADN), which offers great business value for cloud users. A complement to software defined networking (SDN), ADN enables applications to directly control and adapt their networking for optimal performance across both public and private clouds, without compromising on portability or security. With ADN solutions, developers and administrators can automatically instrument, analyze and reconfigure the virtual network of resources to ensure that their cloud-based applications will perform optimally under highly variable conditions, and that they can quickly respond should outages or problems occur.

No Time for Down Time
The development and use of applications is becoming more widespread as businesses and consumers become more reliant on the cloud. However, this growing use of applications is causing serious performance issues because the network infrastructure and the server-centric approach to cloud provisioning was not designed to optimally handle the demands put on it by dynamically changing applications. Among the problems cloud app developers and administrators face are unpredictable latencies and uneven user experiences; disastrous cascading effects from bottlenecks, failures and cloud outages; poor performance and lack of isolation from a large number of users; wasted capacity resulting from over-provisioned infrastructure; and unmanageable network complexity and spiraling costs from lack of visibility in resource interactions.

These issues are among the causes of numerous large-scale cloud outages. No organization, no matter how small or large, can escape the problem. Consider, in mid-March, Google Drive experienced three outages in the course of a week, including one caused by load balancing and latency issues that lasted several hours and impacted at least one-third of its users.[1] But, Google is not alone. Microsoft, Amazon and other prominent cloud providers have been similarly impacted. In a recent study, RightScale reported that there were 27 notable publicly reported cloud outages around the world in 2012, and the average time to recover was 7.5 hours.[2]

Given the extended time that it appears to be taking to recover from outages, it's clear that providers have difficulty quickly and easily identifying where the problems may reside and addressing them effectively. Because businesses are now relying on the cloud for mission-critical applications, any downtime could result in a serious loss of revenue or productivity - not to mention great dissatisfaction for users of these apps and their cloud providers.

Achieving the True Promise of the Cloud
As cloud services have grown in prominence, users' expectations have grown as well. The cloud should enable more flexibility and self-service provisioning, and ensure that applications are readily available whenever and wherever users need to access them. In addition, the cloud should be able to reduce IT complexity and costs.

However, with more cloud solutions available and the industry maturing rapidly, these benefits still tend to be elusive. While costs of public cloud solutions tend to be very low, users are left with the IT headaches of having to manage their own networking and dealing with the complexity of developing, provisioning and supporting their applications. Because of the server-centric approach to cloudsourcing, it's often difficult for app developers and administrators to determine where problems reside. As a result, they spend too much time having to instrument, check, script and reconfigure their infrastructure so it adapts to the changing workloads, instance failures and cloud outages.

The problem will no doubt get worse. As demand for increasingly sophisticated cloud apps grows, the server-centric and black box network infrastructure will make it exponentially harder for developers and app administrators to monitor, troubleshoot and fix problems that may occur.

Tackling the Challenges with ADN
The industry has been taking steps to address these issues, the first of which is the introduction of software-defined networking, or SDN. Defining a new approach to simplify networking, SDN decouples the control plane and the data plane, supporting network virtualization and enabling administrators to quickly provision network connections, rather than having to manually configure policies.

While SDN is a good start, it has limitations because it only focuses on the network and doesn't adequately address many problems that app administrators will continue to experience in the cloud, such as predictability and performance issues. Nor is SDN, as delivered today, meeting the need for infrastructure differentiation required by various apps.

Filling that void is application-defined networking, or ADN, a software solution that resides as an intermediate layer between the network infrastructure and the application. Unlike SDN, which controls forwarding of individual packets, ADN is focused on the overall needs and policy goals of an application, managing entire application glows and accelerating and streamlining the movement of data throughout the entire virtual infrastructure. ADN is designed to give greater visibility into the various pieces that combine to impact infrastructure performance. Another hallmark of ADN is automation. With ADN, apps are able to automatically generate the code to adapt their networking environment, ensuring that they continue to run smoothly and can quickly adjust to constantly changing demand.

Seeing the Benefits of ADN
ADN is designed to leverage any cloud network machine, whether or not it's powered by SDN, to provide software and services that intelligently articulate network resources with the other computing resources. By doing so, ADN enables better communication between components in the cloud, which in turn prevents the cascading effects of unpredictable loads, outages and other related problems.

Among the benefits of ADN are:

  • Reduced Overprovisioning and CapEx Costs: For most application developers and administrators, it's common to overprovision network resources to handle the peak load. In this scenario, users are paying for capacity that they don't need the majority of the time. By making the network more aware of the needs of applications, ADN can help ensure that greater capacity is delivered just in time, and only for as long as required by an application.
  • High Elasticity: Traditional networking technologies are very static in their configuration, and are not designed to be driven by applications. With ADN, developers and administrators can more closely integrate the application with the network resources, and have the infrastructure adjust as the demand for the application changes. This eliminates latency and other problems that may result as users tweak network resources to optimize performance.
  • Simplifies IT: As applications are developed or deployed, there are a number of components, such as nodes, load balancers and servers, that users have to add, remove and otherwise manipulate. Trying to configure the network can be quite complex, particularly for businesses that are trying to benefit from the cloud, but don't have networking expertise in-house. Additionally, if something goes wrong, it's difficult for users to see into the black box and determine exactly what needs to be fixed. ADN strives to eliminate the guesswork, giving users greater visibility into whether the failure is in a router, load balancer, firewall or the application itself.
  • Optimized Performance: Overall, ADN delivers a holistic view of the cloud infrastructure at all layers and how it interacts with applications running over it. By having greater visibility into each component, and the interaction between all components, ADN can enable applications to keep functioning properly, while supporting more proactive troubleshooting and recovery from possible problems.

Cloud infrastructure and computing are becoming an integral part of everyday life - from delivering productivity and cost-savings for businesses, to providing entertainment and enhanced communications between individuals. As a result, performance, uptime and reliability are vital to enabling users to achieve the full promise of the cloud. ADN, in many ways, is a great equalizer - helping users leverage the benefits of the cloud while minimizing the cost and complexity that may have been burdening greater cloud adoption.

References:

  1. http://www.infoworld.com/d/cloud-computing/google-drive-hit-three-outages-week-215073
  2. http://talkincloud.com/cloud-computing-research/cloud-outages-power-loss-blamed-main-cause

More Stories By Pascale Vicat-Blanc

Pascale Vicat-Blanc is founder and CEO of Lyatiss Inc, a software provider for application defined networking (ADN) solutions. She has more than 20 years of international experience on network and cloud computing technologies, having served as research director at INRIA (French National Institute for Research in Computer Science), CIO of a Grid 5000 Data Center, team leader at INRIA-Bell Labs and a project manager with the CERN (European Organization for Nuclear Research).

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


CloudEXPO Stories
Extreme Computing is the ability to leverage highly performant infrastructure and software to accelerate Big Data, machine learning, HPC, and Enterprise applications. High IOPS Storage, low-latency networks, in-memory databases, GPUs and other parallel accelerators are being used to achieve faster results and help businesses make better decisions. In his session at 18th Cloud Expo, Michael O'Neill, Strategic Business Development at NVIDIA, focused on some of the unique ways extreme computing is being used on IBM Cloud, Amazon, and Microsoft Azure and how to gain access to these resources in the cloud... for FREE!
CI/CD is conceptually straightforward, yet often technically intricate to implement since it requires time and opportunities to develop intimate understanding on not only DevOps processes and operations, but likely product integrations with multiple platforms. This session intends to bridge the gap by offering an intense learning experience while witnessing the processes and operations to build from zero to a simple, yet functional CI/CD pipeline integrated with Jenkins, Github, Docker and Azure.
"We do one of the best file systems in the world. We learned how to deal with Big Data many years ago and we implemented this knowledge into our software," explained Jakub Ratajczak, Business Development Manager at MooseFS, in this SYS-CON.tv interview at 20th Cloud Expo, held June 6-8, 2017, at the Javits Center in New York City, NY.
Today, we have more data to manage than ever. We also have better algorithms that help us access our data faster. Cloud is the driving force behind many of the data warehouse advancements we have enjoyed in recent years. But what are the best practices for storing data in the cloud for machine learning and data science applications?
The hierarchical architecture that distributes "compute" within the network specially at the edge can enable new services by harnessing emerging technologies. But Edge-Compute comes at increased cost that needs to be managed and potentially augmented by creative architecture solutions as there will always a catching-up with the capacity demands. Processing power in smartphones has enhanced YoY and there is increasingly spare compute capacity that can be potentially pooled. Uber has successfully been able to harness the excess capacity of privately owned vehicles and turned into a meaningful business. This concept can be step-functioned to harnessing the spare compute capacity of smartphones that can be orchestrated by MEC to provide cloud service at the edge.