We live in a choice-driven society powered by technology. Whether it’s cutting the cable cord and moving to à la carte television or ditching the concept of ownership and opting to take part in the sharing economy, people want to pick and choose what works best for their individual needs instead of relying on a single provider or product.
In the case of cloud computing, a plethora of options is available to IT network teams for how they use infrastructure in private clouds, public clouds, and on-premise data centers and connect enterprise applications running in these environments.
With cloud adoption no longer an “if” but a “how,” companies are focusing on creating the right mix of public and private clouds to maximize efficiency and minimize spending. According to RightScale's State of the Cloud Survey, 85% of enterprises in 2017 are now creating cloud strategies that include multiple clouds, of which 58% are planning to implement hybrid cloud environments.
While many benefits exist to utilizing both private and public clouds, there are some challenges to the hybrid cloud environment that need to be addressed to ensure security and efficiency.
Handling Encrypted Traffic
Understanding the types of security threats (many of which revolve around access) helps address them properly. The rise in hybrid clouds has also resulted in an increase in the amount of encrypted traffic that application networking solutions need to address. Load balancers, which are often tasked with decrypting incoming traffic, need to support the latest encryption protocols and ciphers such as elliptic curve cryptography (ECC). They also need to scale horizontally and support granular per-app services to address the unique needs to each application.
The Risk Of Retrofitting Network Design
Building a hybrid cloud network requires a strategic plan to ensure successful integrations between public and private cloud services, as well as any on-premise applications and data. In today’s reality, many companies find themselves creating a plan after they’ve already started utilizing different providers and clouds. Though it’s not unusual, there are factors to consider to address efficiency and constraints. For example, it’s important to know what applications can be successfully migrated and run in which environments and where a particular set of data is allowed to reside.
Capacity Planning In The Cloud Era
According to the State of the Cloud Survey, the top initiative in 2017 for the majority of all cloud users is optimizing costs. While public clouds can appear to be a low-cost option, the price can jump quickly with heavy usage, especially when utilizing a dynamic cloud that continues to scale as needed. On the flip side, setting up a private cloud is not an insignificant endeavor, given the physical hardware environments needed. One method to curtail costs, which comes up in my conversations with Avi Networks customers, is to look at ways to minimize investment in hardware load balancers where software alternatives are available.
Over the past few years, I have had conversations with network administrators and architects at several large enterprises and found that hardware load balancers are significantly over-provisioned, to the tune of about 80%. One hapless network administrator at a large online retailer confided in me by saying, “I would rather run at 20% capacity for the better part of the year than be caught without being able to handle sudden traffic bursts. The potential loss of business and lack of peace of mind is just not worth it.” Software load balancers architected for cloud-native applications are finally making it possible to expand or shrink load-balancing capacity dynamically in response to real-time traffic needs.
To make the shift to a hybrid cloud computing environment, a few key best practices will help ensure success:
1. Automate Services
Hybrid clouds are designed to thrive on automation. For example, using a next-generation load-balancing solution allows predictive app auto-scaling. Such systems are analytics-driven and can automatically recognize changing traffic patterns in real time and spin up additional instances without human intervention. This end-to-end automation across the environment is made possible when a hybrid cloud traffic management system is in place.
IT services teams can build this self-service infrastructure to not only optimize computing resources and provisions on the fly but also to shift workloads as needed. These types of capabilities provide the agility that hybrid clouds promise with built-in elasticity, responsiveness and efficiency.
Managing cloud services with multiple providers or across environments doesn’t need to be challenging; network teams simply need a single and central point of management across all environments, no matter where applications are running. Because public and private cloud infrastructures operate independently, it’s critical to use technology that provides portability of data and applications between clouds.
For example, when it comes to application networking services, software load balancers that combine central management together with per-app delivery services enable a high degree of customization and flexibility. The alternative of deploying an expensive, monolithic hardware load balancer in front of multiple applications creates problems when each application needs to be maintained or updated, causing downtime for others.
3. Use Vendor-Agnostic Services
Now is the time to take advantage of the healthy competition that is brewing between cloud providers to avoid getting locked into a single cloud provider. Because not all cloud providers deliver consistent services, it behooves companies to remain nimble and test out different services to find the ones that work best.
By keeping the marketplace open and utilizing different providers, companies can take advantage of the myriad options to lower costs and increase performance, especially when they build a hybrid cloud that utilizes the best-of-breed from private and public clouds.
Hybrid cloud computing is growing in popularity because it offers companies flexibility, scalability and agility. To capitalize on this environment, IT teams must spend time creating a strategy that matches their organizational requirements. Putting private and public clouds together requires automation tools and management capabilities to make a system efficient and cost-effective over the long haul.By Ranga Rajagopalan