devopsy.kinsta.cloud

Tag: Distributed Tracing

  • LightStep Unfurls Dynamic Tracing Service

    LightStep Unfurls Dynamic Tracing Service

    LightStep announced a stand-alone dynamic tracing tools capability that visualizes patterns in code based on 100 percent of the data available.

    Company CEO, Ben Sigelman, said LightStep Tracing differs from other approaches to tracing because it does not rely on a small sample of data that usually doesn’t wind up providing much actionable insight. Instead, LightStep Tracing makes it possible to search and visualize distributed traces across 100 percent of the data made available with a software-as-a-service (SaaS) application.

    Unveiled at the QCon London 2019 conference, LightStep Tracing is based on functionality originally developed in the application performance management (APM) platform, dubbed LightStep [x]PM, that is optimized for microservices-based applications. LightStep Tracing, however, does not require DevOps teams to implement LightStep [x]PM to take advantage of distributed tracing.

    Distributed tracing is becoming more widely employed because that capability makes it much simpler to visualize patterns indicative of a potential performance issues. Sigelman built one of the first instances of distributed tracing during his tenure at Google. Known as Dapper, the project now analyzes more than two billion transactions per second. Sigelman subsequently co-created the OpenTracing Application Programming Interface (API) standard that is now made available under the auspices of the Cloud Native Computing Foundation (CNCF).

    According to Sigelman, one of the most common mistakes many DevOps teams make today is overvaluing a distributed trace as a data source. While access to distributed traces are required, the real value comes from being able to dynamically inspect 100 percent of all the traces available in a way that visually surfaces patterns, added Sigelman. DevOps teams should be careful what conclusions they draw from analysis based on samples of data because outliers that could have a significant impact on application performance can be easily missed, cautioned Sigelman.

    LightStep Tracing provides DevOps teams with access to dynamic system diagrams that visually summarize the bottlenecks, as well as real-time performance histograms enable organizations to understand what defines normal application behavior in their environment. Sigelman said that DevOps teams, for the first time, can also capture shareable snapshots of system behavior that explain performance anomalies based on statistical summaries of thousands of relevant traces. Within minutes, Sigelman said LightStep Tracing can be employed to instrument and visualize service-to-service interactions across web, mobile, microservices, serverless and monolithic applications.

    One of the more frustrating aspects of DevOps is that most teams will spend far more time looking for the source of the problem, versus fixing it. But until the source of a problem can be identified, the reputation of the IT organization in the eyes of the application owner continues to suffer. There’s also an unfortunate tendency for DevOps teams to spend an inordinate amount of time in “war rooms” trying to proof that whatever element of the IT environment they manage is not part of the root cause of the problem. Rather than wasting all that time at the expense of the reputation of the IT organization, perhaps the time has come to discover and visualize issues hopefully long before any application owner even becomes aware they exist.

    — Mike Vizard

  • VMware Wavefront Adds AI, Distributed Tracing

    VMware Wavefront Adds AI, Distributed Tracing

    VMware has updated its Wavefront monitoring service to include support for distributed tracing along with anomaly detection capabilities enabled by machine learning algorithms.

    Stela Udovicic, senior director for product marketing for Wavefront by VMware, said the addition of the distributed tracing support, now available as part of a public beta, means that a single monitoring service now can provide metrics, histograms and the tracing function required to provide a holistic view of the IT environment. Those capabilities eliminate the need to employ multiple services to achieve the same goal, she noted.

    In contrast, Wavefront makes it easier to search for traces containing a specific service or application programming interface (API) to determine which ones have errors or long response time. Additionally, DevOps teams can filter the trace search to show traces containing spans from a given cluster or shard.

    At the same time, VMware is adding an AI Genie capability that promises to make it easier to troubleshoot existing needs and identify future application and IT infrastructure requirements, said Udovicic. That data is collected in real time and presented in a way that allows even the most novice users to visualize using analytics tools built into the Wavefront services, added Udovicic.

    Wavefront is based on an agentless architecture that in part makes it possible for the service to process millions of metrics, histograms and traces per second. That capability is especially critical in IT environments that have embraced microservices based on containers that tend to expand across an enterprise IT environment, said Udovicic. Wavefront already supports Kubernetes clusters.

    In general, Udovicic noted that developers tend to embrace a variety of open source approaches to monitoring applications as they build their applications. But a SaaS application such as Wavefront makes it easier for enterprise IT organizations to apply monitoring across a broad range of applications. To facilitate that transition, Wavefront provides an OpenTracing/OpenCensus compliant solution that includes a drop-in replacement for the Zipkin and Jaeger distributed tracing software, which is at the core of many open source monitoring tools.

    VMware is clearly betting that increased reliance on AI will push organizations toward relying more on cloud services. Machine learning algorithms require access to massive amounts of data before they can construct AI models capable of accurately predicting events. The cost of collecting and storing all that data, plus the expertise required to train the AI models, are beyond the capabilities of most DevOps teams.

    Most enterprise IT organizations have applied monitoring tools to only their most critical applications, mainly because of cost concerns. But as applications based on microservices proliferate across the enterprise, it’s becoming apparent that all the dependencies that will exist between various application services will need to be continuously monitored. The only issue now is determining how best to achieve that goal as part of an integrated set of DevOps processes.

    — Mike Vizard

  • New Relic Tracing Software Pinpoints Source of IT Issues

    New Relic Tracing Software Pinpoints Source of IT Issues

    At a FutureStack 2017 conference this week, New Relic previewed an implementation of distributed tracing software that has the potential to close the war rooms where IT operations assign blame for application failure.

    New Relic CEO Lew Cirne says a Distributed Tracing capability that New Relic previewed this week will enable developers or IT operations teams to pinpoint the precise source of any given performance issue. Rather than wasting hours on root cause analysis, Cirne says DevOps teams will now be armed with dashboard they can navigate to visually inspect what elements of a distributed computing environment are specifically affecting their applications.

    Based on the open-source Open Tracing software being advanced by the Cloud Native Computing Foundation (CNCF), Cirne says New Relic opted to embrace an open initiative that should lead to more actionable intelligence being derived from the data New Relic collects.

    At this juncture, Cirne contends that deploying an application in production that has not been instrumented borders on the criminal. He concedes, however, the preponderance of applications deployed in production environments today are not instrumented largely because the total cost of instrumenting legacy applications has been cost-prohibitive. But as IT monitoring software becomes less expensive to invoke using agents that communicate back to a software-as-a-service (SaaS) application, most new applications being deployed in the cloud are now being instrumented.

    To help further that goal, New Relic this week also announced support for .Net Core 2.0 framework recently made available on Microsoft Azure and added support for six additional Amazon Web Services (AWS) services. New Relic now supports a total of 25 AWS services.

    At the conference Cirne also showed how IT organizations can now invoke New Relic application programming interfaces (APIs) to spin up a dashboard to track specific events associated with, for example, the rollout of a new application.

    Access to the data New Relic surfaces, notes Cirne, in turn helps fuel adoption of DevOps because organizations gain confidence in their ability to more rapidly build and deploy applications. That confidence will only increase as IT organizations avail themselves of artificial intelligence (AI) software in the form of machine learning and, soon, deep learning algorithms to make more reliable decisions based on patterns in their IT environments. New Relic, he adds, is able to fuel those algorithms because the service collects 1.5 billion metrics and events every minute. That said, humans will still be required to manage the overall IT environment.

    Cirne maintains that the best way to collect all that data is to employ agents, noting New Relic is adopting those agents to make them lightweight enough to even deploy on microservices based on containers.

    New Relic naturally is not the only provider of IT monitoring tools investing heavily in advanced algorithms. In the not-too-distant future, it’s probable that algorithms deployed by different vendors will share insights with one another. In the meantime, IT administrators can take some comfort in the fact that it should soon become easier to manage IT at scale.

    — Mike Vizard