We’ve been hard at work over the past months to ready our real-time traffic control for AWS’s new service, AWS Gateway Load Balancer (GWLB), and I am excited to announce today the general availability of Glasnostic as an appliance on this new and exciting service. By deploying Glasnostic as an appliance integrated with GWLB, customers with complex and dynamic microservice environments can now, with a single click, avail themselves of Glasnostic’s capability to detect, respond to and proactively manage their unpredictable and disruptive traffic patterns—automatically, in any environment and in real-time. As a result, organizations can manage their digital transformation much more effectively.
Companies increasingly adopt an agile operating model where small, “2-pizza” teams develop in parallel and deploy independently from each other. Microservices are ideally suited to this agile operating model because they allow companies to move fast by reusing existing services and thus rapidly building new functionality on top of existing ones. As a result, though, companies end up operating complex, continually growing and rapidly evolving environments.
These modern environments are fundamentally different from traditional, self-contained applications. While failures in conventional application environments are predominantly due to defects in code, today’s complex and dynamic environments fail predominantly due to unexpected and unpredictable interactions between systems. These interaction behaviors tend to cause large-scale, cascading and highly disruptive failures. Worse, when these problems arise, companies may have difficulty determining what happened and what steps to take to remediate them.
The business’s desire to build and compose services freely and rapidly and the unpredictability of failures and breaches in the resulting complex, dynamic environments mean that operators must be able to detect and respond to the unexpected at runtime and in real-time.
Conventional observability tools are designed to provide deep visibility into the thread of execution of individual transactions. Tracing, debugging and traditional monitoring, however, is ineffective when dealing with the complex and unpredictable service interactions of modern environments.
Glasnostic solves these problems by providing operations and security teams with high-level observability and enabling them to take control of the situation—as it develops, automatically and in real-time. It is a lightweight and non-invasive solution that uses a distributed, integrated fleet of virtual service routers to turn existing networks into an active, application-aware fabric that detects and responds to disruptive behaviors immediately, which dramatically slows any cascading effects.
Glasnostic is also infrastructure-agnostic. It can be either consumed in the cloud or installed on premises. The service routers that form Glasnostic’s data plane deploy as virtual appliances in physical or virtual machine environments, as a DaemonSet in Kubernetes or on top of an existing Envoy-based service mesh such as Istio.
For instance, on Amazon Elastic Compute Cloud (Amazon EC2), this involves deploying a virtual appliance in each VPC and Availability Zone and configuring the local route tables to route traffic to it. Similarly, on Amazon Elastic Kubernetes Service (Amazon EKS), it involves deploying a DaemonSet or, alternatively, a plugin on top of Istio. While none of these steps are difficult by themselves, they require a certain amount of manual installation and configuration.
Glasnostic enables operations and security teams to detect, respond to, and proactively manage disruptions, automatically and in real-time. The new AWS Gateway Load Balancer service brings customers this ability with a single click, thus greatly benefiting their agility.
With GWLB, though, traffic between VPCs is transparently intercepted and handed to the Glasnostic appliance, without any additional configuration. In other words, customers gain a zero-install and zero-configuration path to observability and control of VPC ingress and egress via an Internet Gateway and traffic between VPCs via AWS Transit Gateways. Customers do not need to deploy a virtual appliance for every Availability Zone or configure every VPC to route traffic to Glasnostic. Instead, they merely add a Gateway Load Balancer Endpoint (GWLBE).
Deploying Glasnostic as a GWLB appliance is not only a convenient alternative to a manual install. It also provides several additional benefits, including:
- Allowing operators to focus on more impactful tasks as manual installation steps are eliminated.
- Increased agility as environments can be set up and torn down faster.
- Simplified creation of high-availability configurations.
- Elimination of network configuration.
With GWLB, customers can simply use AWS CloudFormation to install Glasnostic and are instantly ready to observe and control their complex and dynamic environments.
Companies are adopting microservices because of their benefits to business agility, but this new service-based agility comes at the price of increased complexity and an increased rate of change. Because these complex and dynamic environments are prone to unpredictable and disruptive interaction behaviors, operations and security teams need to be able to detect, remediate and manage these threats—anywhere, automatically and in real-time.
GWLB allows customers on AWS to avail themselves of this critical ability with a single click, without additional infrastructure components or additional configuration, thus greatly benefiting their agility.