HPCC Systems vs Tencent Cloud Application Performance Management

HPCC Systems is a free, open-source platform for enterprise big-data processing and data-lake management. Its core components are ECL, a declarative and modular language, Thor for data refinement, and Roxie for data delivery. ECL's compiler optimizes programs for parallel processing and compiles them into C++. Thor handles ingestion, transformation, linking, and indexing; Roxie serves concurrent queries. Clusters can scale from two computers to more than a thousand commodity-hardware nodes. The cloud-native platform runs on Kubernetes and supports Azure Kubernetes Service and Amazon Elastic Kubernetes Service. Storage options include AWS S3, Azure Blob Storage, AWS Elastic Block Store, Azure Files, and Azure Disks. Security capabilities include end-to-end encryption, service meshes, OAuth 2.0 with Azure AD support, JWT, and configurable controls for data and services. The Machine Learning Library offers ECL-accessible algorithms for building and testing prediction models. Integrations include Spark, Kafka, Couchbase, Redis, Memcached, Pentaho, R, JDBC, Java APIs, and Tableau data connectors. ESP exposes queries through XML, HTTP, SOAP, and REST interfaces, and deployed queries can be called with REST/JSON. The platform is free to use under the Apache 2.0 license and is self-hosted.

Tencent Cloud Application Performance Management (APM) monitors application performance through metrics, traces and logs, and supports analysis and troubleshooting across distributed architectures. It uses probes to monitor applications in real time without code modifications. Traces map request paths across microservices and can include request parameters, transaction data, errors, exceptions, method stacks and instance environment details. APM also discovers application dependencies and shows global topologies for examining application and component bottlenecks.

Its analysis groups performance and exception metrics by dimensions such as API, exception and database call, helping identify slow APIs, SQL statements and recurring exceptions. Named database integrations include MySQL, Oracle, MongoDB and Redis. APM supports OpenTelemetry and works with SkyWalking and Jaeger tracing protocols when reporting configuration is changed. It can associate traces with Cloud Log Service logs, display Cloud Monitor dashboards, and send traces to CKafka with configuration and permissions. Applications outside Tencent Cloud can connect when network access is available. New users receive a 15-day trial, limited to 100 million reported Spans and seven-day trace storage; pay-as-you-go also lists one million Spans per day free for reporting and retention. Pricing includes a $0.02 a month pay-as-you-go rate billed hourly per million Spans reported, plus retention at 0.012 USD per million Spans per day.