devopsy.kinsta.cloud

Category: AI

Swimm Adds Generative AI Chat Tool for Documentation

Swimm today announced it has added a chat tool that enables developers to use natural language to surface insights into code.

Company CEO Oren Toledano said the /ask Swimm tool makes it simpler to launch queries that enable developers to better understand how code was constructed.

The /ask Swimm tool aggregates documentation along with other related data to surface factors that are not evident in the code itself, such as documentation of business decisions, product design considerations and decisions concerning why specific architectural choices were made. It automatically captures and updates code-related knowledge to provide a continuous feedback loop as code, documentation, files and repositories are created and updated to provide deeper insights into code that goes beyond what might have been actually documented.

The overall goal is to provide a mechanism that enables developers to understand how and why code was constructed within the context of an integrated development environment (IDE), said Toledano.

Swimm has previously made available a platform that uses generative AI platform to create a static analysis of documentation. That capability should make it simpler for organizations to track documentation at a time when generative AI platforms such as ChatGPT are exponentially increasing the amount of code being written. In theory, those platforms should be able to also create documentation for that code, but there will still be a need for tools to track and analyze it.

In the longer term, Swimm plans to continue to extend its usage of generative AI to provide deeper levels of insights across entire applications and software ecosystems to address that challenge, noted Toledano.

In the short term, however, it’s clear that AI is making developers more productive than ever. In fact, the volume of code being generated might soon overwhelm existing DevOps pipelines and workflows. Most DevOps teams will need to revamp those workflows as the pace at which more applications than ever are being developed and deployed faster than ever.

It’s still early days as far as generative AI adoption is concerned, but it’s clear many developers are already using it to write code. How much of that code makes it into a production environment is difficult to determine, but a lot of that code is going to be of varying quality. A general-purpose AI platform such as ChatGPT was trained using code collected from all across the web. The code generated by these platforms is only as good as the examples used to train it, all of which were created by human developers who may have made mistakes, such as including code that has known vulnerabilities.

Regardless of whether code was generated by a machine or a human, the probability that documentation will need to be analyzed as part of a code review process is always high. The challenge and the opportunity is to determine the best way to apply AI to a longstanding challenge that is now being exacerbated by the existence of AI tools that, with each passing day, are only becoming more accessible to developers.

January 17, 2024
AI a Key Driver Behind HPE’s $14 Billion Deal for Juniper

Hewlett Packard Enterprise is looking to become a more significant player in the networking space through its planned $14 billion acquisition of Juniper, a deal that it hopes will make it a more formidable rival to longtime market leader Cisco Systems.

The deal, announced Tuesday after the markets closed, is a big deal in the early days of the new year for a networking industry that has become central in an IT sector that is becoming more distributed and more cloud-native.

During a virtual briefing with analysts and journalists this morning, HPE CEO Antonio Neri described an HPE centered around its networking business that has AI capabilities and its GreenLake edge-to-cloud platform of IT services at its foundation.

“HPE will be a new company where networking will be the core foundation of everything we do,” Neri said. “We’re going to accelerate what we call an AI-driven agenda, and that will allow us to capture this massive inflection point.”

Even when the deal closes – which is expected to happen later this year or in early 2025 – HPE will still likely be in third place in the global networking space behind Cisco and Huawei, but will have a stronger portfolio that will not only include greater AI capabilities but also a stronger presence in both the enterprise and telecom spaces.

Once it closes, Juniper CEO Rami Rahim will lead the combined HPE networking business and report to Neri.

Juniper’s Mist AI is at the Center

Unsurprisingly, AI was a key component of the deal. In a research note, Will Townsend and Patrick Moorhead, analysts with Moor Insights and Strategy, wrote that their thinking after initial news reports about a possible deal circulated was that HPE likely was looking for a “strong AI anchor” for its portfolio of hardware, software, and GreenLake IT consumption services.

“AI is hot, ignited by the attention being directed toward generative AI, the underlying large language models, and many promising use cases,” Townsend and Moorhead wrote. “One could argue that beyond the AIOps capability found in the HPE Aruba Networking portfolio today, HPE needs further AI depth to remain competitive and continue to grow its top-line revenue and profitability. Juniper could deliver on that front.”

Rahim called AI “the biggest inflection since the dawn of the internet itself” and added that the combination of HPE and Juniper “will be able to bring the depth and the breadth of the portfolios necessary to capture the full market opportunity that AI presents in front of us.”

AI in networking is a strength for Juniper, which in 2019 bought Mist Systems and its AI technologies, including the Mavis virtual network assistant, which the analyst wrote serves “as the tip of the spear for Juniper’s reinvigorated efforts within the enterprise for WLAN, LAN, WAN, and SD-WAN solutions.”

“By all measures, the Mist acquisition has been a success, with Juniper growing its enterprise install base at a faster rate than its service provider business over the last 12 to 18 months,” Townsend and Moorhead wrote.

AI and Networking

AI will play an increasingly important role in networking going forward, from dynamically adjusting bandwidth and self-correcting in the network for maximum uptime to quickly finding root causes for problems and deploying virtual network assistants.

In a blog post last month, Liz Centoni, executive vice president, chief strategy officer, and general manager for applications at Cisco, wrote that according to the company’s AI Readiness Index, 95% of survey respondents have an AI strategy in place or being developed but that only 14% are ready to integrate it into their business.

That will change in 2024, as AI becomes central in everything from networks to infrastructure to applications.

“[Generative AI] will offer APIs, interfaces, and services to access, analyze, and visualize data and insights, becoming pervasive across areas such as project management, software quality and testing, compliance assessments, and recruitment efforts,” Centoni wrote. “As a result, observability for AI will grow.”

With Juniper, HPE is looking to grow its AI capabilities. According to Juniper’s Rahim, his company has spent the past seven years building out its AIOps capabilities, with marked results, including seeing revenue from products that use AI growing almost 100% year-over-year for the last two quarters.

“And we’re starting to see early momentum from our investments in enabling and automating AI data centers, including those handling emerging AI training and inference models,” he wrote in a blog post. “This combination [with HPE] will supercharge our offerings in AI-native networking. Together, we will accelerate innovation at every layer: compute, storage and networking; silicon, systems and software; campus and branch, data center and the wide area network.”

GreenLake is a Foundational Part

Neri, in 2019, declared that within three years, HPE would be offering its entire portfolio as a service through the GreenLake platform, which has expanded significantly over the past several years. During the briefing, Neri and Rahim focused on GreenLake as the base for building out a layered AI-driven architecture at a massive scale and what the HPE CEO called the most complete networking portfolio.

“What we’re doing today with this proposed acquisition is to create a new core,” he said. “The core is the network fabric by which we deliver hybrid cloud and AI-native services with the rest of the portfolio we have. We deliver all this unified experience to the same platform called HPE Greenlake, which is a true edge-to-cloud platform. … The core business inside GreenLake – by which we deliver the entire set of services, whether it’s cloud-native or AI-native – is networking.”

The Juniper deal will give HPE the ability to deliver a better experience to users, which is important given that the most difficult part of the architecture is the network, Neri said, adding, “That’s why we believe this changes dramatically who we are and it will increase our relevancy with customers and partners to ways we couldn’t imagine before.”

There’s More to the Deal

Beyond AI, acquiring Juniper will enable HPE to play a greater role in a networking space that has long been dominated by Cisco. According to reports, Neri told HPE employees in a letter that the deal would enable his company to essentially double the size of its networking business and expand significantly in such areas as the enterprise, service providers, cloud environments and – again – AI.

HPE has been growing its networking business for a number of years, in large part through acquisitions, including such companies as ConteXtream and Plexxi. It also acquired the high-performance Slingshot interconnect through its 2019 acquisition of supercomputer maker Cray. In the data center, its most significant purchases were Aruba Networks in 2015 – which has become the core of its networking business – and Silver Peak five years later to bolster its software-defined networking (SDN) chops.

That said, HPE’s networking revenue is about $3.7 billion – a fraction of Cisco’s $24.1 billion – according to Zeus Kerravala, principal analyst with ZK Research. Bringing Juniper – and its 11,0000 employees – into the fold will bolster that to about $5.8 billion, making HPE the number-three vendor in terms of revenue. Still a long way off from Cisco, but HPE would become a stronger rival with more capabilities.

Moor’s Townsend and Moorhead wrote that while there are significant benefits in the merger, there also is a large amount of product overlap that will need to be addressed.

“Serious roadmap rationalization will need to occur, and without a doubt, some solutions will be sunsetted,” they wrote. “There will also likely be a reduction in force as enterprise and service provider networking product teams and internal channel sales and marketing personnel are blended.”

January 10, 2024
From Reaction to Robots: Riding the AI Wave in 2024

As we navigate another year of consistent zero-day breaches, legislative pivots, the explosion of AI tooling and threat actors growing bolder and more desperate, it’s safe to say that getting comfortable with change is a requirement for thriving in the technology industry.

We occupy a notoriously unpredictable space, but that’s half the fun. Compared to many other verticals, technology—especially cybersecurity—is relatively youthful, and the future should be something we can all look forward to blossoming in sophistication alongside the technology we swear to protect.

So, what can we expect in the industry in 2024? We put our heads together, looked into our crystal ball, and these were the results:

Government Regulations Around AI Will Turn the Industry Upside Down

It was the talk of the conference circuit in 2023, with several high-profile presentations at Black Hat, DEF CON, Infosecurity Europe and many more warning of the explosive changes we can expect from AI implementation across every industry, especially cybersecurity. As tends to happen with low barriers to entry for such transformative technology, adoption has outpaced any official regulation or mandates at the government level.

With significant movements in general cybersecurity guidelines and benchmarks around the world, including CISA’s Secure-by-Design and -Default principles in the U.S. and similar initiatives from the UK and Australian governments, it is essentially a foregone conclusion that regulations around AI use will be announced sooner rather than later.

While much of the debate surrounding the mainstream use of AI tooling and LLMs has centered around copyright issues with training data, another perspective delves into how AI is best used in cybersecurity practices. When it comes to coding, perhaps its most human quality is its similar hardship in displaying contextual security awareness, and this factor is deeply concerning as more developers are adopting AI coding assistants in the construction of software. This has not gone unnoticed, and in a time of increased scrutiny for software vendors adopting security best practices, government-level intervention certainly would not surprise.

… And Demand for AI/ML Coding Tools Will Create a Need for More Developers, not Less!

Much has been written about the AI takeover, and for the better part of a year, we have been subject to a plethora of clickbait headlines that spell doom and destruction for just about every white-collar profession out there, and developers were not spared.

After months of speculation and experimentation with LLMs in a coding context, we remain entirely unconvinced that development jobs are at collective risk. There is no doubt that AI/ML coding tools represent a new era of powerful assistive technology for developers, but they are trained on human-created input and data, and that has rendered the results far from perfect. Perhaps if every developer on the planet was a top-tier, security-minded engineer, we might see genuine cause for concern.

However, just as the average adult driver vastly overshoots their ability (notice how everyone says they’re a great driver, and it’s always other people who lack skill? That’s a classic example of the Dunning-Kruger effect!), so too does the development community, especially when it comes to security best practices. According to one Stanford study into developer use of AI tooling, it is likely that unskilled developers using this technology will become dangerous. The study claimed that participants who had access to AI assistants were more likely to introduce security vulnerabilities for the majority of programming tasks yet also more likely to rate their insecure answers as secure. This poses a significant issue; poor developers will be enabled to introduce security issues faster, and if anything, this will only increase the need for security-skilled developers with the knowledge and expertise to code securely and use AI technology safely.

We Will see Consequences for Software Vendors who Don’t Ship Secure Code

CISA director Jen Easterly has made it abundantly clear that software vendors should not be permitted to “pass the buck” when it comes to security within their products, highlighting that the current responsibility for software safety is largely passed to the consumer.

We have agreed with this view, and we believe it will take a shakeup of this magnitude to move the needle toward code-level security—not to mention appropriately educating developers—to be taken seriously.

While CISA’s powers only extend so far—essentially, they can only enforce secure-by-design practices to vendors that sell to federal agencies—this still presents a new security benchmark to hit for many large software vendors. Colonial Pipeline, SUNBURST and, more recently, the MOVEit data breach are all large-scale cyberattacks that affected government-level systems at some point, and with these new guidelines in play, there is a real possibility that future highly visible breaches will receive greater scrutiny and reprimand.

Reactive Security Will Start to be Seen as Old-School

As the goal of increased cyber resilience continues to dominate cyber strategies across multiple verticals, those who rely on reaction and incident response as the only core tenets of their plan will find themselves in a place of unacceptable exposure and risk.

Security professionals must act swiftly in the face of adversity and outright attack, but modern times call for modern solutions, and we simply cannot afford to take a less-than-holistic approach. “Shift left” needs to be more than a rapidly aging buzzword; code-level security should be prioritized, alongside upskilling and verifying the competence of the developers working on the software and critical digital infrastructure we take for granted. After all, who wants the security program equivalent of an Etch-a-Sketch when they could have an iPad Pro?

Now, more than ever, governments and enterprises alike must commit themselves to a preventative, high-awareness security program in which every member of staff is enabled to share responsibility. It’s not enough to cite the cybersecurity skills gap as a reason for falling behind; investment in security-aware developers and fostering collaboration between them and their AppSec counterparts should be a driving force in remaining as secure as possible, both as an organization and within the production of software.

Matias Madou, co-founder and CTO at Secure Code Warrior, contributed to this article.

January 10, 2024
Broadcom Survey Surfaces Raft of IT Automation Challenges

A survey of 501 IT leaders published today by Broadcom suggested that islands of automation that operate independently of each other are major contributing factors to organizations failing to meet service level agreements (SLAs).

The survey found that 98% of respondents encountered at least one automation issue, with 61% noting that SLA issues are a monthly occurrence. More than a third (35%) said these issues occur weekly. A full 88% indicated SLA breaches directly impacted customers in a significant way.

The survey also found nearly three-quarters (74%) lacked visibility across their data center and cloud environments, with an equal number lacking end-to-end visibility of their automated business processes. Only 28% possessed tools that could predict automation issues, and more than two-thirds (68%) were unable to model or predict whether a new or updated automation job would actually meet SLAs. A full 69% said they are unable to properly size automation workload resources, with 61% having no way to simulate automation workloads before implementation, the survey found.

Only slightly more than half (55%) of respondents, however, could provide SLAs and key metrics in a format that is appropriate for business stakeholders.

Aline Gerew, head of automation for the Agile Operations Division at Broadcom, said that while it’s routine for IT teams to leverage a range of automation tools and platforms to manage IT, it’s clear they are challenged to manage those platforms in a comprehensive fashion. As a result, whenever there is an issue with one tool or platform, it creates a bottleneck that impacts customer experience in a way that ultimately adversely impacts revenue, she noted.

In fact, many IT teams are now starting to identify automation architects within their IT teams to enable DevOps teams in a more holistic manner, added Gerew.

Overall, the survey found that 80% of respondents are using three or more automation platforms, with just over a quarter (27%) using six or more.

Nearly three quarters (74%) are also relying on three or more observability solutions, with just under a quarter (24%) using six or more. Just under three-quarters of respondents noted they are consequently subjected to alert storms that ultimately serve to make it more challenging to resolve issues. More than half (56%) of respondents are blind to the downstream effects of an initial alert and, as a result, do not inform response teams of which systems, jobs, applications and business processes are affected.

In some ways, IT automation has become too much of a good thing. Every provider of an IT platform makes automation tools available, but as they are invoked, each one creates yet another island that IT teams are expected to manage in isolation. However, all it takes for an SLA not to be met is for an issue to arise that IT teams can’t easily determine the root cause of because one or more automation platforms obscure the underlying issue in the absence of any standardization, noted Gerew.

Given the complexity of modern IT environments, automation is now a requirement. The issue, as always, is finding a way to apply it all in the most transparent way possible.

January 9, 2024
The Age of Platform Engineering: What to Expect in 2024

Platform engineering has been gaining traction over the last few years, with some likening it to the next iteration of DevOps. Gartner has called it one of the top strategic trends of 2024 – and has predicted that 80% of large software engineering organizations will establish platform engineering teams by 2026. But what’s driving the momentum?

I believe it’s a result of maturity in the adoption of the cloud by enterprises of all sizes and shapes. As more and more enterprises started using the cloud and embraced DevOps as a path towards quickly taking features to the market, there was a clear need to rethink how the underlying infrastructure and platforms were used for enterprise applications. Platform engineering emerged as the new path to building and managing the underlying application infrastructure.

It’s been a growth year for platform engineering, but in today’s era of hybrid-cloud-meets-generative-AI, we see new trends emerging in this space. So, what’s next? Here is what we expect to see as we head into 2024.

Product-Centric Funding Model

Traditionally, IT was seen as a gatekeeper and cost center. With the cloud, developers gained more leverage in procuring the infrastructure they wanted for their applications. Enterprises realized the value of modernizing their IT operations to meet this modern need and wanted to transform their cloud operations to be part of the innovation team. They see platform engineering as the transformational approach that aligns cloud operations with developers and business users. However, platform engineering requires a product-centric funding model instead of project-centric models in the past, as a product-centric model doesn’t have a start and end date like projects and aligns closely with the product lifecycle with ongoing funding. Enterprises need to recognize this shift from the status quo widely, and we expect this realization to come for businesses in 2024.

Platform Engineering as an Extension to DevOps

As enterprises embraced DevOps to accelerate their application development, they still used the traditional gatekeeping models with cloud operations. The embrace of traditional guardrails in the cloud created friction and slowed innovation. With its “as code” model, platform engineering is well suited for reducing friction and accelerating innovation. We expect organizations to embrace platform engineering as the natural extension of DevOps to operate at the speed of business.

Generative AI Capabilities

Generative AI is at the peak of the hype cycle as we close out 2023, and in 2024, we will see some enterprise-ready use cases come to life. Generative AI will augment platform engineering with code assistants to create infrastructure-as-code, governance-as-code and security-as-code to ensure smooth Day 2 operations. We will also see chatbots leveraging generative AI to help SREs and other platform engineering pros with documentation, infrastructure playbooks and other relevant use cases.

2024 is a critical year for organizations as they fend off the economic forces and embrace technological advancements to compete in the global market. Most CXOs will see platform engineering as the top strategic trend to focus on because it brings productivity and cost efficiencies to their organization. This is only the beginning, and the coming decade will be defined by platform engineering, laying a firm foundation for AI-driven innovation in enterprises.

January 9, 2024
Best of 2023: Copilots For Everyone: Microsoft Brings Copilots to the Masses

As we close out 2023, we at DevOps.com wanted to highlight the most popular articles of the year. Following is the latest in our series of the Best of 2023.

Microsoft has been doing a lot to extend the coding ‘copilot’ concept into new areas. And at its Build 2023 conference, Microsoft leadership unveiled new capabilities in Azure AI Studio that will empower individual developers to create copilots of their own. This news is exciting, as it will enable engineers to craft copilots that are more knowledgeable about specific domains.

Below, we’ll cover some of the major points from the Microsoft Build keynote from Tuesday, May 23, 2023, and explore what the announcement means for developers. We’ll examine the copilot stack and consider why you might want to build copilots of your own.

What is Copilot?

A copilot is an artificial intelligence tool that assists you with cognitive tasks. To date, the idea of a copilot has been mostly associated with GitHub Copilot, which debuted in late 2021 to bring real-time auto-suggestions right into your code editor.

“GitHub Copilot was the first solution that we built using the new transformational large language models developed by OpenAI, and Copilot provides an AI pair programmer that works with all popular programming languages and dramatically accelerates your productivity,” said Scott Guthrie, executive vice president at Microsoft.

However, Microsoft recently launched Copilot X, powered by GPT-4 models. A newer feature also offers chat functionality with GitHub Copilot Chat to accept prompts in natural language.

But the Copilot craze hasn’t stopped there—Microsoft is actively integrating Copilot into other areas, like Windows and even Microsoft 365. This means end users can write natural language prompts to spin up documents across the Microsoft suite of Word, Teams, PowerPoint and other applications. Microsoft has also built Dynamics 365 Copilot, Power Platform Copilot, Security Copilot, Nuance and Bing. With this momentum, it’s easy to imagine copilots for many other development environments.

Having built out these copilots, Microsoft began to see commonalities between them. This led to the creation of a common framework for copilot construction built on Azure AI. At Build, Microsoft unveiled how developers can use this framework to build out their own copilots.

Building Your Own Copilot

Foundational AI models are powerful, but they can’t do everything. One limitation is that they often lack access to real-time context and private data. One way to get around this is by extending models through plugins with REST API endpoints to grab context for the tasks at hand. With Azure, this could be accomplished by building a ChatGPT plugin inside VS Code and GitHub Codespaces to help connect apps and data to AI. But you can also take this further by creating copilots of your own and even leveraging bespoke LLMs.

Understanding The Azure Copilot Stack

Part of the Azure OpenAI service is the new Azure AI Studio. This service enables developers to combine AI models like ChatGPT and GPT-4 with their own data. This could be used to build copilot experiences that are more intelligent and contextually aware. Users can tap into an open source LLM, Azure OpenAI or bring their own AI model. The next step is creating a “meta-prompt” that provides a role for how the copilot should function.

So, what’s the process like? Well, first, you create a new Azure OpenAI resource. Then you add a data source, which can be as simple as a PDF file or Word document. Next, you can then choose to have the generative AI augment responses or limit itself to the supplied data. Then, the service is exposable through an API to be injected into any application format. The result could be a chat assistant that loops into private data and works well with your own apps and data.

The copilot stack consists of the frontend design, an orchestration layer and a foundational AI model at its core.

Microsoft has also built tools to keep private data safe and encrypted throughout the AI orchestration process. “AI orchestration involves grounding, prompt design and engineering, evaluation and AI safety. These are kind of the core fundamentals for how you create great copilot experiences,” said Guthrie.

You Get a Copilot, and You Get a Copilot and …

The surge of copilots comes at a time when many companies are integrating LLMs and generative AI into their offerings. These powerful AI additions have the ability to accelerate low-code development with advanced auto-complete features. And the Azure OpenAI service looks to be a promising platform to quickly spin up custom copilots.

“AI is going to profoundly change how we work and how every organization operates, and every existing app is going to be reinvented with AI, and we’re going to see new apps be built with AI that weren’t possible before,” said Guthrie.

By constructing their own copilots, organizations will continue to customize the experience with generative AI and give it more context into specific domains to solve unique pain points. And integrating copilots into the realm of end-user interfaces will help open up accessibility to AI in a broader sense.

“Software has eaten the world; now it’s AI’s turn,” said Thomas Dohmke, CEO, GitHub.

January 4, 2024
Skillsoft Survey Sees AI Driving Increased Need to Retrain IT Teams

More organizations than ever will need to invest in IT training as advances in artificial intelligence (AI) transform roles and responsibilities in the coming year.

A survey of 2,740 IT decision-makers conducted by Skillsoft, a provider of an online training platform, finds two-thirds (66%) were already dealing with skills gaps in 2023. As AI becomes more pervasively applied to the management of IT, that skills gap is only going to widen given the limited pool of IT professionals that have any experience using AI to manage IT.

Skillsoft CIO Orla Daly said it’s already apparent AI creates an imperative for training because there are simply not enough IT people with the requisite skills required. In fact, the survey finds nearly half of IT decision-makers (45%) plan to close skills gaps by training their existing teams. That training is crucial because the primary reason IT staff change jobs is a lack of growth and development opportunities, noted Daly.

While there is naturally a lot of consternation over the potential elimination of IT jobs, in the final analysis, AI will add more different types of jobs than it eliminates, adds Daly. Each of those jobs will require new skills that will need to be acquired and honed, she notes. “Training is the price of innovation,” said Daly.

In the meantime, there is much interest in finding ways to automate existing IT processes to create more time for IT teams to experiment with AI technologies, said Daly.

The report also finds well over half of IT decision-makers (56%) expect their IT budgets to increase to help pay for new platforms and tools, compared with only 12% expecting a decrease.

It’s not clear to what degree AI will transform the management of AI, but it’s already apparent that many manual tasks involving, for example, generating reports are about to be automated. The instant summarization capabilities that generative AI enables also promise to dramatically reduce the time required to onboard new members to an incident response team. Rather than having to allocate someone on the team to bring new members up to speed, each new member of the team will use queries framed in natural language to determine for themselves the extent of the crisis at hand.

In addition, many tasks that today require expensive specialists to perform might become more accessible to a wider range of organizations as, for example, more DevOps processes are automated. That doesn’t necessarily mean that DevOps as an IT discipline disappears as much as it leads to the democratization of best DevOps practices.

Each IT organization in the year ahead will need to determine to what degree to rely on AI to manage IT processes. It may take a while before IT teams have enough confidence in AI to rely on it to manage mission-critical applications but many of the tasks that today conspire to make the management of IT tedious will undoubtedly fade away. The challenge and the opportunity now is identify those tasks today with an eye toward revamping how IT might be managed in the age of AI tomorrow.

January 2, 2024
Best of 2023: Microservices Sucks — Amazon Goes Back to Basics

As we close out 2023, we at DevOps.com wanted to highlight the most popular articles of the year. Following is the latest in our series of the Best of 2023.

Welcome to The Long View—where we peruse the news of the week and strip it to the essentials. Let’s work out what really matters.

This week: Amazon Prime Video has ditched its use of microservices-cum-serverless, reverting to a traditional, monolithic architecture. It vastly improved the workload’s cost and scalability. (more…)

December 28, 2023
Best of 2023: Will ChatGPT Replace Developers?

As we close out 2023, we at DevOps.com wanted to highlight the most popular articles of the year. Following is the latest in our series of the Best of 2023.

AI is buzzing again thanks to the recent release of ChatGPT, a natural language chatbot that people are using to write emails, poems, song lyrics and college essays. Early adopters have even used it to write Python code, as well as to reverse engineer shellcode and rewrite it in C. ChatGPT has sparked hope among people eager for the arrival of practical applications of AI, but it also begs the question of whether it will displace writers and developers in the same way robots and computers have replaced some cashiers, assembly-line workers and, perhaps in the future, taxi drivers.

It’s hard to say how sophisticated the AI text-creation capabilities will be in the future as the technology ingests more and more examples of our online writing. But I see it having very limited capabilities for programming. If anything, it could end up being just another tool in the developer’s kit to handle tasks that don’t take the critical thinking skills software engineers bring to the table.

ChatGPT has impressed a lot of people because it does a good job of simulating human conversation and sounding knowledgeable. Developed by OpenAI, the creator of the popular text-to-image AI engine DALL-E, it is powered by a large language model trained on voluminous amounts of text scraped from the internet, including code repositories. It uses algorithms to analyze the text and humans fine-tune the training of the system to respond to user questions with full sentences that sound like they were written by a human.

But ChatGPT has flaws—and the same limitations that hamper its use for writing content also render it unreliable for creating code. Because it’s based on data, not human intelligence, its sentences can sound coherent but fail to provide critically informed responses. It also repurposes offensive content like hate speech. Answers may sound reasonable but can be highly inaccurate. For example, when asked which of two numbers, 1,000 and 1,062, was larger, ChatGPT will confidently respond with a fully reasoned response that 1,000 is larger.

OpenAI’s website provides an example of using ChatGPT to help debug code. The responses are generated from prior code and lack the capability to replicate human-based QA, which means it can generate code that has errors and bugs. OpenAI acknowledged that ChatGPT “sometimes writes plausible-sounding but incorrect or nonsensical answers.” This is why it should not be used directly in the production of any programs.

The lack of reliability is already creating problems for the developer community. Stack Overflow, a question-and-answer website coders use to write and troubleshoot code, temporarily banned its use, saying there was such a huge volume of responses generated by ChatGPT that it couldn’t keep up with quality control, which is done by humans. “Overall, because the average rate of getting correct answers from ChatGPT is too low, the posting of answers created by ChatGPT is substantially harmful to the site and to users who are asking or looking for correct answers.”

Coding errors aside, because ChatGPT—like all machine learning tools—is trained on data that suits its outcome (in this case, a textual nature), it lacks the ability to understand the human context of computing to do programming well. Software engineers need to understand the intended purpose of the software they’re creating and the people who will be using it. Good software can’t be built by cobbling together programs with regurgitated code.

For example, ChatGPT cannot understand the ambiguity in simple requirements. Although it is clear that if one ball just bounces and comes back but another bounces and then bounces again, the second ball has traveled further, ChatGPT struggles with this nuance; that nuance will be needed if these systems are ever to take over from developers.

It also has trouble with basic math, such as when it’s asked to determine which is greater and offered a choice between a negative and positive number. ChatGPT confidently tells us a correct summation of the space, but cannot understand that -5 is less than 4. Imagine your thermostats going haywire because the heating kicks on at 40 degrees Celsius instead of at -5 degrees Celsius because the AI program coded it that way!

Pre-trained AI code generation also raises some legal questions with regard to intellectual property rights; it cannot currently distinguish between code that is licensed in a restrictive or open fashion. This could expose people to licensing compliance risk if the AI borrows a prewritten line of code from a copyrighted repository. The problem has already prompted a class action lawsuit against a different OpenAI-based product called GitHub Copilot.

We need humans to create the software people rely on, but that’s not to say there couldn’t be a place for AI in software development. Just like automation is being used by security operations centers for scanning, monitoring and basic incident response, AI could serve as a programming tool for handling lower-level tasks.

This is already happening, to an extent. GitHub Copilot allows developers to use ChatGPT to improve their code, add tests and find bugs. Amazon offers CodeWhisperer, a machine language-powered tool designed to help increase developer productivity using code recommendations generated by natural language comments and code in the integrated environment. And someone has created a Visual Studio code extension that works with ChatGPT.

And one company is testing AI for developers. DeepMind, which shares a parent company with Google, released its own code generation tool, dubbed AlphaCode, earlier this year. DeepMind published the results from simulated evaluations in competitions on the Codeforces platform in Science magazine earlier this month under the headline Machine Learning systems can program too. Headline grammar aside, AlphaCode achieved an estimated rank within the top 54% of participants by solving problems “that require a combination of critical thinking, logic, algorithms, coding, and natural language understanding.” The abstract for the paper says: “The development of such coding platforms could have a huge impact on programmers’ productivity. It may even change the culture of programming by shifting human work to formulating problems, with machine learning being … responsible for generating and executing codes.”

Machine learning systems are becoming increasingly advanced each day; however, they cannot think like the human brain does. This has been the case for the past 40+ years of study into artificial intelligence. While these systems can recognize patterns and increase productivity for simple tasks, they may not always produce code as well as humans. Before we let computers do code generation en masse, we should probably see systems like AlphaCode rank in the top 75% of participants on a platform like Codeforces, though I fear this may be too much for such a system. In the meantime, machine learning can help with simple programming problems in the future, allowing developers of tomorrow to think of more complex issues.

At this point, ChatGPT won’t be disrupting any field of technology, especially not software engineering. Concern about robots displacing programmers is vastly overstated. There will always be tasks that developers with human cognition can do that machines will never be capable of.

December 27, 2023
All Your IT Team Wants This Holiday Season is a Break!

The holiday season is all about giving. As organizations increasingly look to IT as they move toward new digital tools and processes, now is the perfect time to give back to IT teams tirelessly working to keep the modern enterprise online. Whether your system performance has been naughty or nice this year, there’s no denying that tech professionals have earned our appreciation, respect—and the tools to set them up for success in 2024.

For IT teams limited in both time and resources, simply maintaining systems can feel as impossible as squeezing themselves down a chimney or delivering gifts to millions of homes in a single night. On top of that, instead of being greeted with milk and cookies, they’re inundated with endless performance issues, support requests and alerts—leaving little time left over for the important work of innovating.

They say the best gifts are the ones you can’t wrap. That holds true for IT teams, too. This year, bring your organization the gift of a simpler, speedier, more rewarding workload. If your team is dreaming of a tech-savvy future, here are some enterprise software solutions to make their lives easier that they won’t want to re-gift:

Enjoy the View With Observability

Everyone loves to cozy up at home during a winter snowstorm, but with the widespread migration to combined remote, on-premises and distributed hybrid environments, the daily monitoring journey for today’s IT teams is more akin to trekking blindly through a blizzard.

Observability tools are metaphorical snowshoes and goggles that can help them not only weather the storm but see clearly from the mountaintop. Observability is the answer to the modern enterprise’s struggle to gain full visibility into their organization’s apps, networks, databases and infrastructure—something nearly half of IT professionals lack, according to SolarWinds research. IT teams will be able to rest easier at night with visions of sugarplums, rather than outages or anomalies, dancing in their heads.

Even better, integrating artificial intelligence (AI) capabilities into observability solutions to collect and provide data on what’s not performing as expected and why will help your teams take a proactive approach to solving issues.

Lend a Helping Hand With AIOps

AI isn’t just the shiny new toy of the tech world. Organizations using AI for IT operations (AIOps) can give the gift of support to their overworked IT teams by automating some of the time-consuming and mundane tasks that stand between them and a focus on innovation.

Adding AIOps to observability can provide IT teams with maximum visibility into the state of their digital ecosystems through automated discovery and dependency mapping. Additionally, your teams can gain the ability to easily track inbound connections linked across the organization’s application stack and storage volumes with auto-instrumented views.

Today, it simply isn’t feasible for humans alone to manage modern IT environments without intelligent automation. Think of AIOps as a workshop of elves operating in the background to ensure workloads and processes are streamlined and moving as efficiently as possible. With AIOps in place to analyze data and streamline workloads and processes, IT teams are relieved of some pressure—and can focus on accelerating your digital transformation rather than just maintaining it.

Give the Gift of Time

Finally, although you can’t outright give the gift of time to your IT team, you can still arm them with tools that save time year-round. In 2024, more organizations will embrace IT service management (ITSM) to improve their IT outcomes, service delivery and employee satisfaction.

SolarWinds research found that enterprises using AI-powered ITSM tools were able to reduce system downtime by 21% and decrease time spent resolving incident and service requests by 23%. In fact, the average SolarWinds customer reported savings of 23 hours weekly due to reduced ticket volume—nearly the equivalent of adding an employee to the team.

This holiday season, the best gifts you can give your IT teams are the actionable, predictive intelligence and tools they need to optimize performance, cut down on complications and quickly resolve issues. (As a bonus, it will certainly make you more popular than the guy from The 12 Days of Christmas who thought it was a good idea to gift his partner geese, milkmaids and leaping lords.)

December 22, 2023