Microservice application performance management provider, Instana, has released new capabilities for monitoring the VMware vSphere Suite and the applications running on vSphere infrastructure.
Instana correlates infrastructure and application performance metrics. This release includes the ability to discover, map and monitor components running on VMware's vSphere suite. As with other Instana supported infrastructure components, application performance metrics are analysed along with the new vSphere metrics. Chris Farrell, technical director and APM strategist at Instana, said:
Application migration is one particular use case for which Instana's broad infrastructure and architectural support combine to add value. Whether migrating from monolith to microservices, physical to virtual hosts, or private to hybrid clouds, Instana's automated discovery and performance monitoring provides a way to capture and compare different deployment options.
Formerly known as VMware Infrastructure, vSphere consists of several technologies that provide the infrastructure to virtualise environments including:
- ESXi – a Type-1 hypervisor that runs on hosts to manage the execution of the virtual machines and allocate resources
- vCenter Server – server management software that provides a centralised platform for controlling vSphere environments
- vCenter Client – a management interface that enables users to remotely connect to vCenter Server
- Virtual Machine File System (VMFS) – a cluster file system for ESXi virtual machines
InfoQ asked Jon Skog, senior product manager at Instana, some questions about the announcement.
InfoQ: Virtual machines or containers? Which is best and why do some organisations have a mix? How does Instana help build resilience into this sort of hybrid environment?
Jon Skog: The decision as to whether to go virtual machine or container is ultimately going to come down to the goals and resources in each application delivery and development organisation. We see a mix of environments at some customers, usually based on how the applications they're operating were built. Instana provides service discovery and application monitoring in all environments so that operations teams can make infrastructure decisions without worrying about their monitoring tool.
InfoQ: How does Instana help if people are migrating from one platform architecture to another?
Instana is also helpful for migration projects - both infrastructure (like physical to virtual) and application architecture (monolith to microservices). The solution's automatic discovery and monitoring, coupled with the ability to compare snapshots pre- and post-migration, make it possible to know that a migration has achieved the desired performance levels.
InfoQ: What kind of actionable data does Instana provide for IT executives and leaders?
Skog: Instana understands that not all events are created equal, and so it only sends actionable alerts when service is impacted. With a combination of anomaly detection, dependency mapping, event correlation, and expert knowledge, Instana is able to achieve automatic root cause analysis of a service impact and delivers the root cause to the end user, so they can take action to resolve.
Pipeline feedback provides immediate data to DevOps teams and developers, giving them feedback to help them determine how new releases and changes are performing. With this feedback, they are able to act on whether to roll back a release, speed up the next release to kill a bug, or to allocate more resources where necessary. Instana also provides users the ability to analyse each trace, call and microservice to uncover bottlenecks and/or performance optimisation opportunities.
Combining this Unbounded Analytics capability with our Application Perspectives feature enables users to to jump from their application data into analytics of application request traces (in context) to get the information to troubleshoot an issue.
InfoQ: Can teams use Instana to compare the cost and performance of different types of application and infrastructure environments?
Skog: Instana automatically discovers all application building blocks: physical components (datacenter, hosts, containers, processes, clusters, etc.), logical components (services, endpoints, applications, traces, and calls), and business components (business services and business processes).
By doing so, Instana makes it possible to see the impact of different environments/infrastructures on your application performance. Instana also makes it possible to see the performance of applications before and after changes - so the impact each change/release has to overall performance is visible.
While Instana does not currently compare the cost of different types of environments, it is possible to see the impact on resources and resource allocations. This data can be used to derive insights into which environments cost more/less over time.
InfoQ: If a team is working in a cloud environment where immutable infrastructure is spun up and down as "cattle not pets", what’s the best practice to ensure Instana agents are deployed where they need to be?
Skog: This depends on the technologies in use. For Kubernetes based environments, Instana offers a Kubernetes Operator that detects applications, services, processes, nodes, hosts, clusters, namespaces, deployments, Kubernetes services, and pods regardless of how quickly they are spun up/down. Instana can also be installed as a DaemonSet, sitting on the outside and reaching into the different components to detect new technologies as they are spun up and/or down.
InfoQ: Is there any kind of infrastructure that Instana can't monitor? What about on-premise ZSeries mainframe?
Skog: Instana is capable of monitoring any type of infrastructure. It doesn't matter if teams are using legacy systems or the most cutting edge technologies, Instana is capable of monitoring from cloud to on-premises to hybrid, up to and including ZSeries mainframes through our partnership with mainstorconcept's z/IRIS technology.
InfoQ: How can a team measure the real business value of a change deployed that Instana detects?
Skog: Instana has a feature called Pipeline Feedback. Pipeline feedback provides data to DevOps teams and developers helping them determine how new releases and changes are performing. With this feedback, they are able to act on whether to roll back a release, speed up the next release to kill a bug, or to allocate more resources where necessary. Additionally, they can dig into data from before a change/release and compare that to the data after a change/release was deployed. Instana uses release markers to show when a release/change has taken place.
InfoQ: How does Instana flag and advise on scaling issues for VMs or containers?
Skog: Instana alerts on various types of contention: transaction times, CPU contention and memory contention, among others. Instana creates an incident based on these factors.
Read more about the new capabilities for monitoring the VMware vSphere Suite and the applications running on vSphere infrastructure
