Apache Kafka stands as a well known open supply occasion retailer and stream processing platform. It has developed into the de facto customary for knowledge streaming, as over 80% of Fortune 500 corporations use it. All main cloud suppliers present managed knowledge streaming companies to satisfy this rising demand.
One key benefit of choosing managed Kafka companies is the delegation of accountability for dealer and operational metrics, permitting customers to focus solely on metrics particular to purposes. On this article, Product Supervisor Uche Nwankwo offers steering on a set of producer and client metrics that prospects ought to monitor for optimum efficiency.
With Kafka, monitoring sometimes includes varied metrics which are associated to subjects, partitions, brokers and client teams. Customary Kafka metrics embody info on throughput, latency, replication and disk utilization. Seek advice from the Kafka documentation and related monitoring instruments to grasp the precise metrics out there on your model of Kafka and how you can interpret them successfully.
Why is it necessary to observe Kafka shoppers?
Monitoring your IBM® Occasion Streams for IBM Cloud® occasion is essential to make sure optimum performance and total well being of your knowledge pipeline. Monitoring your Kafka shoppers helps to determine early indicators of software failure, reminiscent of excessive useful resource utilization and lagging shoppers and bottlenecks. Figuring out these warning indicators early allows proactive response to potential points that reduce downtime and forestall any disruption to enterprise operations.
Kafka shoppers (producers and shoppers) have their very own set of metrics to observe their efficiency and well being. As well as, the Occasion Streams service helps a wealthy set of metrics produced by the server. For extra info, see Monitoring Event Streams metrics by using IBM Cloud Monitoring.
Shopper metrics to observe
Producer metrics
| Metric | Description |
| File-error-rate | This metric measures the common per-second variety of data despatched that resulted in errors. A excessive (or a rise in) record-error-rate may point out a loss in knowledge or knowledge not being processed as anticipated. All these results may compromise the integrity of the info you might be processing and storing in Kafka. Monitoring this metric helps to make sure that knowledge being despatched by producers is precisely and reliably recorded in your Kafka subjects. |
| Request-latency-avg | That is the common latency for every produce request in ms. A rise in latency impacts efficiency and may sign a difficulty. Measuring the request-latency-avg metric can assist to determine bottlenecks inside your occasion. For a lot of purposes, low latency is essential to make sure a high-quality person expertise and a spike in request-latency-avg may point out that you’re reaching the boundaries of your provisioned occasion. You’ll be able to repair the difficulty by altering your producer settings, for instance, by batching or scaling your plan to optimize efficiency. |
| Byte-rate | The typical variety of bytes despatched per second for a subject is a measure of your throughput. If you happen to stream knowledge usually, a drop in throughput can point out an anomaly in your Kafka occasion. The Occasion Streams Enterprise plan begins from 150MB-per-second break up one-to-one between ingress and egress, and you will need to know the way a lot of that you’re consuming for efficient capability planning. Don’t go above two-thirds of the utmost throughput, to account for the doable influence of operational actions, reminiscent of inside updates or failure modes (for instance, the lack of an availability zone). |
Scroll to view full desk
Shopper metrics
| Metric | Description |
| Fetch-rate fetch-size-avg | The variety of fetch requests per second (fetch-rate) and the common variety of bytes fetched per request (fetch-size-avg) are key indicators for a way nicely your Kafka shoppers are performing. A excessive fetch-rate may sign inefficiency, particularly over a small variety of messages, because it means inadequate (probably no) knowledge is being acquired every time. The fetch-rate and fetch-size-avg are affected by three settings: fetch.min.bytes, fetch.max.bytes and fetch.max.wait.ms. Tune these settings to realize the specified total latency, whereas minimizing the variety of fetch requests and probably the load on the dealer CPU. Monitoring and optimizing each metrics ensures that you’re processing knowledge effectively for present and future workloads. |
| Commit-latency-avg | This metric measures the common time between a dedicated document being despatched and the commit response being acquired. Much like the request-latency-avg as a producer metric, a secure commit-latency-avg implies that your offset commits occur in a well timed method. A high-commit latency may point out issues throughout the client that forestall it from committing offsets shortly, which instantly impacts the reliability of knowledge processing. It’d result in duplicate processing of messages if a client should restart and reprocess messages from a beforehand uncommitted offset. A high-commit latency additionally means spending extra time in administrative operations than precise message processing. This concern may result in backlogs of messages ready to be processed, particularly in high-volume environments. |
| Bytes-consumed-rate | This can be a consumer-fetch metric that measures the common variety of bytes consumed per second. Much like the byte-rate as a producer metric, this must be a secure and anticipated metric. A sudden change within the anticipated pattern of the bytes-consumed-rate may symbolize a difficulty along with your purposes. A low charge is perhaps a sign of effectivity in knowledge fetches or over-provisioned sources. A better charge may overwhelm the shoppers’ processing functionality and thus require scaling, creating extra shoppers to stability out the load or altering client configurations, reminiscent of fetch sizes. |
| Rebalance-rate-per-hour | The variety of group rebalances participated per hour. Rebalancing happens each time there’s a new client or when a client leaves the group and causes a delay in processing. This occurs as a result of partitions are reassigned making Kafka shoppers much less environment friendly if there are loads of rebalances per hour. A better rebalance charge per hour may be brought on by misconfigurations resulting in unstable client habits. This rebalancing act may cause a rise in latency and may lead to purposes crashing. Be certain that your client teams are secure by monitoring a low and secure rebalance-rate-per-hour. |
Scroll to view full desk
The metrics ought to cowl all kinds of purposes and use circumstances. Occasion Streams on IBM Cloud present a wealthy set of metrics which are documented right here and can present additional helpful insights relying on the area of your software. Take the following step. Study extra about Event Streams for IBM Cloud.
What’s subsequent?
You’ve now received the data on important Kafka shoppers to observe. You’re invited to place these factors into apply and check out the absolutely managed Kafka providing on IBM Cloud. For any challenges in arrange, see the Getting Started Guide and FAQs.
Learn more about Kafka and its use cases
Provision an instance of Event Streams on IBM Cloud
Was this text useful?
SureNo
