October 2022 Summaries

Hash functions, invented by Hans Peter Luhn in 1953, are crucial in software engineering for mapping keys to binary numbers, predominantly used in hash tables for constant time lookups. They are classified into cryptographic hash functions, used for security purposes like password verification and file integrity, and non-cryptographic hash functions, facilitating statistical methods and probabilistic data structures. The text discusses the development of "hash4j," a new Java library addressing limitations in existing libraries like Guava and Zero-Allocation-Hashing, offering high-quality, fast, non-cryptographic hash functions such as Wyhash and Komihash. Hash4j provides features like a streaming interface for direct object hashing, automatic handling of variable-length fields, and support for ordered and unordered collections, minimizing hash collisions through innovative strategies. The library, open-sourced on GitHub, ensures reliability with extensive testing and a 100% branch test coverage policy, using VarHandle for fast and future-proof memory access. Future enhancements include advanced data structure implementations and similarity hashing techniques for efficient data processing.

Companies like BT are leveraging AIOps and digital transformation to modernize their operations, enhance efficiency, and improve customer experiences as they expand their service portfolios and move to the cloud. This approach involves not only digitization but also automation through software intelligence, allowing organizations to better manage cloud complexity and data profusion. BT, with its substantial global presence, utilizes Dynatrace's observability platform to address these challenges, enabling the company to reduce digital incidents by 50% and mean time to resolution by 40%, translating to significant internal value and return on investment. By proactively identifying and addressing issues with AI-powered tools, BT can prioritize and resolve problems before they impact users, illustrating the critical role of AIOps in managing the increasingly complex IT environments of today’s digital landscape.

Open source software (OSS) is gaining prominence in cloud computing as it offers a collaborative, efficient development approach that is replacing proprietary-only code bases, allowing organizations to focus on building their own intellectual property. According to the Open Source Initiative, 95% of organizations increased or maintained their use of OSS in 2023, with significant growth driven by the need to accelerate innovation and reduce costs. OSS encourages industry consensus around technology standards, fostering collaboration among competitors to solve common problems, as demonstrated by successful projects like OpenTelemetry, which addresses telemetry data tracing across distributed systems. While vulnerabilities exist in all software, OSS benefits from a strong community that can quickly identify and address security issues. Dynatrace, among others, supports OSS as it aligns with industry standards and promotes a culture of innovation, predicting a shift towards open-source solutions as proprietary-only software becomes less viable.

NoOps, or "no operations," is a concept that seeks to automate IT processes to the extent that a dedicated operations team becomes unnecessary, evolving from the DevOps model where developers and operations work collaboratively. Initially introduced by Forrester in 2010, NoOps didn't gain much traction due to the complexities involved in evolving DevOps practices. However, the rapid digital transformation prompted by the COVID-19 pandemic has reignited interest in NoOps, as organizations face resource constraints and increased pressures on IT teams. The idea is not to eliminate operations staff entirely but to reduce their burden by automating routine tasks, allowing them to focus on innovation. AIOps, which combines big data and machine learning, has been crucial in this shift, offering advanced capabilities for automating IT operations processes like anomaly detection and root cause analysis. Modern AIOps solutions, such as those offered by Dynatrace, further enhance this approach by providing real-time data analytics and intelligent problem-solving across hybrid and multi-cloud environments. As organizations prepare for a future with diverse cloud infrastructures, NoOps is becoming increasingly feasible, supported by advancements in cloud automation and AIOps, promising a streamlined IT operations landscape.

Dynatrace's Davis AI engine enhances site reliability engineers' (SREs) capabilities by streamlining exploratory and proactive analyses, ultimately saving time and reducing complexity in managing cloud-native environments. By automatically analyzing thousands of signals, Davis identifies anomalies and provides explanations, allowing SREs to understand and address the root causes of unexpected changes without the need for custom dashboards. This capability is particularly beneficial in complex scenarios such as Kubernetes environments, where introducing technologies like service meshes can lead to issues like "zombie" pods, which consume resources unnecessarily. Davis enables efficient diagnosis and remediation of such issues by providing detailed insights and suggesting best practices, such as using Kubernetes' activeDeadlineSeconds spec, concurrencyPolicy, and namespace quotas, to prevent resource exhaustion. The new exploratory analysis feature, part of Dynatrace SaaS version 1.254, is set to be released in November and will extend its utility beyond Kubernetes to other domain-specific analysis pages, promising to enhance the operational efficiency and innovation potential for SRE and DevOps teams.

Dynatrace's SaaS release version 1.253 introduces several enhancements and new features aimed at improving observability, data management, and security across IT infrastructures. The release includes the integration of Grail, a data lakehouse optimized for Dynatrace Davis® AI that simplifies data storage and querying with the new Dynatrace Query Language (DQL), allowing for exploratory analysis of event and time series data without predefined schemas. Log Management and Business Observability are enhanced with real-time data analysis capabilities, ensuring precision without data loss, and sensitive data masking features are introduced to protect user information. Additionally, new configurations for metric events support anomaly detection and auto-adaptive thresholds, while usability improvements in Data Explorer and API changes enhance data interaction and visualization. The release also addresses vulnerability detection in language runtimes and resolves several technical issues to ensure seamless operations and user experience.

Earlier this year, Dynatrace announced its participation in the open-source project OpenFeature, which has since been accepted as a sandbox project by the Cloud Native Computing Foundation (CNCF). OpenFeature, backed by leading feature flag companies, provides a vendor-neutral specification and software development kits (SDKs) for various programming languages, facilitating the rapid release and testing of software features. Feature flags, which act as runtime toggles for application functionality, have become crucial for organizations aiming to balance speed with software quality, enabling agile development practices such as hypothesis-driven development and experimentation. OpenFeature offers a vendor-agnostic API that promotes interoperability, allowing teams to seamlessly integrate feature flagging with existing or new systems, and fostering a community of feature flag experts. For observability, OpenFeature's integration with Dynatrace's distributed tracing technology, PurePath®, allows teams to monitor feature flag impacts and make informed, data-driven decisions. The collaboration between Dynatrace and OpenFeature promises to enhance observability in feature flagging, with future plans to integrate with Dynatrace's Grail™ data lakehouse and Davis AI, making it an opportune time for organizations to explore this synergy at events like KubeCon 2022.

A high-severity remote code execution (RCE) vulnerability, known as CVE-2022-42889 or Text4Shell, was disclosed on October 13, 2022, affecting the Apache Commons Text library, which provides text processing functions. Although some perceive it as the successor to Log4Shell, its impact is considered less severe because it is harder to exploit and not as widespread. The vulnerability impacts versions 1.5 to 1.9 of the library but has been addressed in version 1.10.0. Organizations are advised to upgrade to this latest version to mitigate risks. Various tools and strategies can be employed to detect if an organization is affected, including Dynatrace Application Security and the Maven dependency tree. The vulnerability allows for execution of expressions, DNS resolution, and loading values from remote servers through string interpolation, which makes certain features like "script," "dns," and "url" potentially exploitable. Despite its high NVD severity rating of 9.8, other agencies have rated it as less severe due to its limited adoption and exploitation difficulty.

AIOps, or Artificial Intelligence for IT Operations, leverages machine learning and AI to automate IT processes, enhancing efficiency and observability by managing the vast data generated in complex, multicloud environments. It addresses incident management through event correlation, anomaly detection, and causality determination, offering two main approaches: traditional correlation-based and modern deterministic, causal AI. While traditional AIOps reduces alerts but struggles with scalability and real-time insights, modern AIOps integrates deterministic AI for precise, continuous monitoring, facilitating dynamic cloud operations and full-stack observability. This modern approach enables organizations to manage complex interdependent microservices across multiple clouds, driving efficiency, innovation, and enhanced business outcomes by reducing manual data analysis and alert fatigue. Key capabilities of effective AIOps solutions include comprehensive integration with existing systems, real-time, continuous insights, topology mapping, and distributed tracing, particularly within environments like Kubernetes, where reliability and scalability are crucial. Deterministic AI in modern AIOps offers advanced analytics and automation, fostering proactive problem resolution and improved IT operations, enabling IT teams to deliver better user experiences.

Dynatrace Davis® AI offers automatic analysis of IT infrastructure, identifying abnormal situations, their impact, and root causes while supporting various observability use cases without needing configuration. The introduction of "metric events" allows for domain-specific customization, enabling users to observe metrics, select appropriate monitoring strategies, and create event triggers for anomalies. Davis processes billions of events to autonomously detect and analyze root causes, but businesses can tailor it to detect critical anomalies relevant to them. The system features two anomaly detectors: the metric key for monitoring raw data with static thresholds, and the metric selector for analyzing historical data with strategies like auto-adaptive thresholds. Users can configure and preview alert settings, adjust the sensitivity of alerts, and customize event messages with contextual information. Dynatrace version 1.253 will automatically enable these features across SaaS environments, encouraging user feedback for continuous improvement.

Dynatrace introduces configuration-as-code for logs, enhancing scalability in enterprise environments by automating log data acquisition, filtering, masking, and anonymizing. This system reduces manual work for DevOps and SRE teams by centralizing the orchestration of OneAgents, which dynamically tie logs to distributed traces, user sessions, and IT topology for enriched observability. The platform addresses the challenges of ephemeral log sources in Kubernetes-based architectures and varying team requirements by offering flexible configuration rules at multiple levels, ensuring compliance through on-source data filtering and sensitive data masking. Dynatrace's log acquisition toolkit automatically discovers relevant log sources, allowing customized setups that align with specific business needs, thereby streamlining observability efforts and reducing complexity in large-scale environments.

The OneAgent 1.251 release, launched on October 14, 2022, introduces several updates and enhancements across various platforms, including Java, Node.js, and PHP, along with operating system support changes. Key features include default installation of the Npcap driver on Windows from version 1.255, which may cause network disruptions on certain Windows versions without specific hotfixes. The release also signals the end of support for older iOS and Xcode versions, aligning with Apple's guidelines for app submissions. Notably, the release includes multiple cumulative updates addressing issues such as input validation vulnerabilities in Real User Monitoring (RUM) and improvements in OneAgent's support for various Java versions and frameworks. Additionally, several resolved issues improve the stability and functionality of the OneAgent components, including configuration syncing for the iOS Session Replay module and enhanced stability for the .NET code module following Microsoft security updates.

The Dynatrace Managed release notes for version 1.252, published on October 12, 2022, detail several critical updates and improvements. This release marks the end of support for certain operating systems, such as CentOS/RHEL/OL 8.2 and others, with version 1.258 being the last upgradeable version on these platforms. The Davis Assistant feature is scheduled to be discontinued by January 31, 2023, affecting integrations with platforms like Slack and Microsoft Teams. Key enhancements include a predictable data retention management system that automatically adjusts storage in case of insufficient disk space and the automatic update of threat intelligence feeds in the Application Security module. The release also upgrades JRE for cluster node components, improves SAML-based SSO configuration, and enhances email validation for user accounts. Several specific issues have been resolved, including memory leaks, display errors, API issues, and a vulnerability in the Apache Commons Text library. The release notes also outline future changes in operating system support, extending into 2026.

As organizations increasingly shift towards serverless computing to capitalize on its benefits, such as dynamic scalability, cost-effectiveness, increased agility, faster time to market, reduced latency, resource optimization, and minimal infrastructure maintenance, they must also navigate associated challenges. These challenges include loss of control, security concerns, architectural complexity, third-party dependency, and difficulties in testing and monitoring serverless environments. Despite these drawbacks, serverless computing remains attractive due to its on-demand access to back-end resources, particularly for code development and execution. To effectively balance these benefits and challenges, organizations are encouraged to employ AI-driven intelligent observability tools, such as Dynatrace PurePath and the Dynatrace Davis AI engine, which can provide end-to-end visibility, pinpoint root causes of issues, and offer solutions to maintain optimal serverless operations.

Hiring a Site Reliability Engineer (SRE) is a crucial step in advancing an organization's DevOps automation, with their initial focus being on managing the complexities of multicloud environments through advanced observability platforms. The use of AI-driven solutions, such as Dynatrace Cloud Automation, is emphasized for automating delivery and operations, providing real-time insights throughout the CI/CD pipeline. Dynatrace offers a variety of automation solutions, including release automation, service level objectives (SLO) automation, and release validation dashboards, all designed to improve software delivery quality and reliability without altering deployment configurations. These solutions are complemented by dashboards for monitoring key services and signals, aiding in the visualization and analysis of service dependencies and customer impacts. The effective implementation of these solutions can lead to reliable releases, enhancing customer satisfaction and minimizing revenue loss.

Cybersecurity Awareness Month, a collaborative initiative by the U.S. federal government and the IT security industry, seeks to heighten global awareness of cybersecurity's critical importance, particularly as cyberattacks rise amid more complex application deliveries. The 2022 theme, "See Yourself in Cyber," emphasizes the human aspect of cybersecurity, aiming to educate individuals and organizations on making informed online decisions. Key recommendations for enhancing cybersecurity include enabling multi-factor authentication, using strong passwords, recognizing and reporting phishing, and updating software. The initiative underscores the value of integrating security into every phase of the software development lifecycle through practices like vulnerability management, application security monitoring, and DevSecOps automation. Emphasizing a unified, AI-driven observability platform such as Dynatrace, it highlights how these practices can enhance application performance, streamline security processes, and boost overall productivity and efficiency in DevSecOps teams. By adopting robust cybersecurity measures and involving all relevant teams, organizations can mitigate risks, protect sensitive data, and maintain their brand reputation in the face of evolving cyber threats.

Organizations are increasingly turning to data lakehouses combined with software intelligence to transform log data into meaningful insights, advancing IT automation and overcoming the challenges posed by traditional data silos. Logs, distributed traces, and metrics are essential for optimizing application performance and infrastructure, but their value diminishes in the fragmented, multicloud environments common today. A data lakehouse, integrating the flexible storage of a data lake with the management capabilities of a data warehouse, provides a centralized repository for structured and unstructured data, enabling real-time access and analysis without the need for costly indexing. By utilizing AI-driven software intelligence, teams can contextualize and monitor data at scale, enhancing AIOps and providing precise answers to IT issues in real time. This modern approach not only reduces storage costs but also eliminates team silos, offering faster, higher-quality insights and ensuring proactive problem resolution. Dynatrace's unified platform exemplifies these innovations by providing a comprehensive data management solution that enhances cloud transformation and operational efficiency, ultimately delivering better business outcomes.

Dynatrace has introduced Grail, a new data lakehouse technology designed to enhance the capabilities of its Software Intelligence Platform by addressing the challenges posed by the growing complexity and volume of data in multicloud and cloud-native environments. Grail aims to provide boundless observability, security, and business analytics by combining the benefits of data warehouses and data lakes, offering rich data management and analytics features on low-cost cloud storage. It supports the ingestion and processing of vast amounts of data through a scalable cloud-native architecture, allowing for seamless data retrieval and real-time analytics without traditional indexing constraints. Grail's schema-on-read approach provides greater flexibility by allowing data to be stored in its native format, enabling any query at any time. This technology empowers enterprises to drive automation and gain precise insights by leveraging its high-performance analytics capabilities and the Dynatrace Query Language (DQL). Initially available for Dynatrace SaaS on AWS, Grail is expected to expand to other platforms, including Azure, in early 2024.

Dynatrace introduces Grail, a revolutionary data lakehouse technology that enhances log management and analytics by providing context-rich observability, business, and security insights through massively parallel processing. Grail eliminates traditional indexing and storage constraints by enabling schema-on-read queries for precise, value-driven analysis without prior indexing or archiving. Designed to address the inefficiencies of current log management solutions, Grail offers seamless data ingestion across multicloud environments and supports comprehensive analytics using the Dynatrace Query Language (DQL). This approach ensures full data availability, reduces data anxiety, and facilitates the conversion of logs into actionable insights. Dynatrace's integration of Grail with its Software Intelligence platform aims to resolve the challenges of big data management in IT environments, providing a single source of truth and enhancing operational efficiency. Grail is set to become available for Dynatrace SaaS on AWS and Azure, with the potential for further use cases beyond traditional log management and analytics.

Modern organizations require a new approach to software intelligence that leverages data and analytics to unify observability and security data, generating real-time insights crucial for maintaining software performance and operational efficiency. As companies increasingly transform into software entities, they face the complexities of cloud-native environments, which demand seamless functioning of numerous interdependent services. To address these challenges, observability data—comprising logs, metrics, and traces—must be centralized and contextualized to provide actionable insights. An innovative approach, exemplified by Grail, a purpose-built data lakehouse, converges real-time, historical, and predictive analytics on a single platform, offering schema-less data ingestion and direct analytics access without data rehydration. This facilitates precise answers and automated root-cause analysis, enabling IT professionals to make informed decisions without compromising on speed, scale, or cost. The integration of AI further enhances this process by analyzing unified data sets to detect anomalies and prioritize issues, thereby optimizing operations and security. Dynatrace's platform, powered by Grail, exemplifies this integration by eliminating data silos, automating processes, and supporting cloud transformation, ultimately empowering organizations to achieve better business outcomes through tailored observability, AIOps, and application security.