Grafana Labs’ Phlare, Faro to simplify profiling, app observability

Grafana Labs is adding two new products—Phlare and Faro—to its observability stack to optimize infrastructure profiling and application observability.

Roman Samborskyi / Shutterstock

In the face of economic headwinds, as enterprises look to balance a cautious approach to IT expenditures with efforts to achieve operational excellence, open-source data visualization platform provider Grafana Labs on Wednesday said that it was adding two new products—Phlare and Faro—to its observability stack to optimize infrastructure profiling and application observability.

Grafana Labs offers an observability suite via its managed Grafana Cloud and self-managed Grafana Enterprise Stack—both featuring scalable metrics (Grafana Mimir), logs (Grafana Loki), and traces (Grafana Tempo).

In addition to the new Phlare application profiling database and the Faro web front-end monitoring software, the company also announced updates to Mimir, Loki and Tempo.

Observability, which gained more prominence during the pandemic as enterprises tried to shift their operations to the cloud, is a relatively new term in IT used to describe the task of monitoring enterprise applications, data flow and distributed infrastructure.

Systems that offer observability go beyond prior application performance monitoring (APM) programs, offering a high-level overview of IT infrastructure as well as granular metrics, to allow for efficient application, network, data, and security management.

According to a recent report released by APM company New Relic, more than 94% of 1,300 respondents across companies in 16 countries said that observability, sometimes known as  O11y, is critical for the success of their role. In addition, more than 80% of C-suite executives who participated in the survey said they expect to increase their observability budget.

According to a research report released by log-management application provider LogDNA, however, 75% of responding companies are still struggling to achieve true observability despite substantial investments in tools.

The study, which polled 200 senior engineering professionals across the US, showed that two-thirds of organizations currently spend $100,000 or more annually on observability tools, with 38% spending $300,000 or more annually.

Grafana Phlare aims to optimize performance, cost

To help enterprises understand the resource usage of a program or application, which in turn could help optimize performance and cost, Grafana is now offering a continuous profiling database, dubbed Phlare.

Profiling, which can be described as a variation of dynamic code analysis, is generally used to collect performance data for deeper analysis to sort out any issues in terms of query latency or process latency.

“Information about resource usage is collected at regular intervals across an entire compute infrastructure and stored as time series data that can be queried for deeper analysis,” the company said in a statement.

The rationale is to be able to get deeper insights into IT overhead with changing IT environments, the company said, adding that continuous profiling could be considered as the fourth pillar of observability after traces, logs and metrics.

The profiling database, according to the company, will allow enterprise users to visualize profile data alongside metrics, logs and traces, as well as data from disparate data sources.

Faro checks on web-application health

In order to help enterprises monitor the health of their web applications, Grafana Labs has also introduced Faro, an open-source project that enables enterprise users to collect data about the health of the front-end of their web applications.

Faro, which comes in the form of a configurable web software development kit (SDK), can be used to capture observability signals, the company said, adding that these telemetry signals can be used to correlate with the back end of the application and its infrastructure data.

Faro, which is currently in private beta, will be offered under the company’s managed observability offering, Grafana Cloud.

Updates to the Loki, Grafana, Tempo, Mirmir stack

Grafana Lab’s Mirmir, which is an offering geared toward capturing and visualizing metrics for an IT environment within an enterprise, has been updated to support data ingestion from third-party observability offerings such as Influx, Datadog, Graphite as well as  OpenTelemetry metrics, the company said, adding that data can be queried using PromQL—the Prometheus Querying Language, used for the open-source Prometheus time series database. 

In an update to its logs offering, Loki, Grafana Labs also said that it was replacing the old format with a new design based on Prometheus. The new format, which will be made generally available in the upcoming 2.7 release, takes up 75% less space on disk, can be accessed more efficiently, and enables further parallelized query execution, the company said.

In addition to the new format, Grafana Labs will allow enterprises to store logs by moving them to an object storage in AWS, Azure or GCP for long-term storage in order to meet compliance or regulatory guidelines.

These logs can be queried by using LogCLI, a command line interface, or by running the individual enterprise’s Loki instance without additional cost to re-ingest data to Grafana Cloud, the company said.

Grafana combines performance testing, observability