Tag

Strategies

Browsing

The new year is up and running, set to bring solutions and challenges affecting all industries. As the faltering economy continues to revolve around the ravages of broken supply chains and deteriorating cyber security, businesses and analysts alike are turning their attention to what’s next.

TechNewsWorld spoke with IT executives to gather predictions for what will happen in 2023. He offered insightful writing on the wall about what to expect going forward.

One of the most important areas is the need for more effective security to protect cyber infrastructure. Politics aside, Executive Order 14028 issued in May 2021 clarified the priorities. President Biden’s order requires agencies to improve their security to secure the integrity of the software supply chain.

“Software vendors can no longer hide from their shortcomings, and software users can no longer hide from their responsibilities,” said John Geter, chief product and technology officer at RKVST, a SaaS platform for tracking supply chain issues. Technewsworld.

With still a way to go, he sees the digital supply chain finally being recognized as just as important as the physical one. Gator also sees a critical need for suppliers to provide quality and for consumers to control their own risk.

He offered, “Companies and governments around the world are waking up to the fact that the software they use to run their enterprise operations and the hardware and software solutions they use and deliver to customers are represent a significant risk.”

Core Technologies Top Priority

Geter said the current political and macroeconomic conditions are worse than most people predicted, and this is having a chilling effect on innovation.

People will focus more on cost cutting and efficiency. However, this should not diminish the importance of the key technologies being developed.

“But it changes the emphasis from new use cases like proactive cyber defense to improvements in existing use cases like more efficient audits,” he said.


Geter suggested that most supply chain problems come from mistakes or oversights that originate in the supply chain itself, and that leave targets open to traditional cyber attacks.

“It is a subtle distinction but an important one. I believe the bulk of the discoveries arising from improvements in supply chain visibility [in 2023] Will highlight that most threats arise from mistake, not malice,” said Geter.

Year of AI and ML

The new year will bring a renewed focus on machine learning operations (MLOps), predicted Moses Gutman, CEO and co-founder of ClearML, an MLOps platform. It is important to take stock of how machine learning has evolved as a discipline, technology and industry.

He expects artificial intelligence and machine learning spending to continue to grow as companies look for ways to optimize increased investment and ensure value, especially in a challenging macroeconomic environment.

“We’ve seen a lot of top technology companies announce layoffs in late 2022. It’s likely that none of these companies are laying off their most talented machine learning personnel,” Gutman suggested to TechNewsWorld.

However, to make up for the shortfall of fewer people in deep technical teams, companies will need to lean even further into automation to maintain productivity and ensure projects get completed. He also expects to see companies that use ML technology put in place more systems to monitor and conduct performance and make more data-driven decisions on how to manage ML or data science teams.

“With clearly defined goals, these technical teams will need to be more key performance indicator-focused, so leadership can have a more in-depth understanding of machine learning’s ROI. Gone are the days of vague benchmarks for ML,” Gutman said. .

end of talent hoarding

Artificial intelligence and machine learning have become common in the last decade. Those working with ML are likely the most recent employees, as opposed to employees who have been working with AI for a long time.

Many big tech companies started hiring these types of workers because they could handle the financial cost and keep them away from competitors — not necessarily because they were needed, Gutman said.


“From this perspective, it is not surprising to see so many ML workers being laid off given the surplus within large companies. However, with the era of ML talent hoarding coming to an end, it could usher in a new wave of innovation and opportunity for startups,” he observed.

With so much talent now looking for work, he expects to see many displaced workers move out of big tech and into small and medium-sized businesses or startups.

Cloud Predictions

Drew Firmant, vice president of enterprise strategies at Pluralsight, believes that fundamental cloud computing skills will continue to be the most relevant and in-demand worker needs for 2023. This is despite ML and AI getting the most attention.

According to Pluralsight’s State of the Cloud report, 75% of tech leaders are building all new products and features in the cloud. Yet he noted that only 8% of technologists have significant cloud-related skills and experience.

Ironically, lower-level cloud infrastructure skills will continue to be in high demand because using those technologies successfully requires more people than higher-level services, said Mattias Andersen, Pluralsight’s lead developer advocate.

“For example, many organizations now want to own and manage their own Kubernetes clusters, allowing them to hire for Kubernetes administration skills while they offload to a cloud provider,” Anderson told TechNewsWorld. “

tech talent shift

Firmant said an expected shift from consumers of talent to creators of talent will be a key differentiator for cloud leaders in 2023. Gartner reports that 50% of enterprise cloud migrations will be delayed by two years or more due to cloud skills shortages – directly impacting the ability of enterprises to achieve cloud maturity and achieve a return on their technology investment.

“To address the challenges of cloud adoption, enterprises must invest in migrating their talent to the cloud as much as they are investing in migrating their applications,” Firmant told TechNewsWorld. “Lift-and-shift migration strategies limit the benefits of cloud platforms, and the approach doesn’t work well for workforce transformation.”

He urged that in order to achieve a sustainable transformation towards cloud adoption and maturity, enterprises need to invest strategically in skill development programs designed to achieve cloud adoption at critical mass.

Multi-Cloud Adoption

Avoiding vendor lock-in is an important target for 2023. According to Anderson, this is the strategy that is now prevalent across the industry landscape. More enterprises are adopting multi-cloud, either by design or by accident.

“The increase in multi-cloud adoption will accelerate demand for the tools needed to manage the increased complexity as enterprises struggle to reduce their implementation timelines. The trifecta of multi-cloud challenges and solutions in 2023 will include security, cost and operations,” said Anderson.


This, he said, would force another need on multi-cloud strategies. Technologists must become multilingual between two or more cloud providers.

He predicted, “With the current shortage of cloud talent, the multi-cloud strategy trend is expected to add further stress to the existing skills gap.”

open-source role

The focus on ML operations, management and governance will force MLOPS teams to do more with less. According to Gutman, businesses will adopt more off-the-shelf solutions because they are less expensive to produce, require less research time, and can be customized to meet most needs.

“MLOps teams will need to consider open-source infrastructure rather than being locked into long-term contracts with cloud providers. While organizations doing ML at hyper-scale can certainly benefit from integration with their cloud providers, it forces these companies to work the way the provider wants them to work,” he added. Explained.

This means users may not be able to do what they want the way you want, he warned. This also puts users at the mercy of the cloud provider for cost escalations and upgrades.

On the other hand, open source provides flexible customization, cost savings, and efficiency. Users can even modify the open-source code to make sure it works exactly as they want.

Gutman concluded, “especially with teams shrinking in technology, it’s becoming a more viable option.”

Thanks largely to Nvidia, GPUs have almost eclipsed CPUs as a technology driving advancement. When it comes to AI, especially on the training side, GPUs reign supreme. In general, GPUs make our games and even the most advanced photo and video editing tools work better — and with social media, many people are becoming creation experts.

But while AMD, Nvidia, and, more recently, Intel have played in this highly competitive space, all three have come to market recently with distinct strategies, each with unique benefits, risks, and opportunities. As we here in the US recover from our Thanksgiving food coma, let’s explore these strategies.

Then we’ll close with our product of the week: Microsoft’s new Arm-based Surface Pro 9 that could point to the near future of PC advancement.

Marketing Strategies: The Apple Example

My first degree was in merchandising, which deals with branding, shelving, product placement, and product strategy. As a result, I’m a little fascinated with how different companies approach the PC market.

For example, Apple is old school, vertically integrated, and uses lock-in to mine its base for money. This strategy resulted in massive profits, but the company was no longer able to lead the market with innovation. Granted, public companies are measured by profits, not innovation, and if you had to choose one over the other as a CEO who wants to keep your job, profits would also be your focus.

Except for the Microsoft Surface, other PC companies are more liberal, have higher interoperability, no lock-in, and enjoy far more innovation but face less customer loyalty and smaller margins. But they are also less likely to fail catastrophically because they are not mining their customers for money.

IBM’s near collapse in the early 1990s was largely due to a strategy similar to that of Apple. This happened because the clients usually do not like to stake money or do mining for money for extreme profits. Eventually, a company using this approach goes too far, and its customers leave the platform. This is roughly what happened with IBM and it prejudged the strategy Apple still uses.

For now, the strategy still works for Apple. However, this probably won’t work for Dell, HP, or Lenovo, because none of those companies can afford the Microsoft part of the solution individually and because their enterprise buyers are aware of the tactics used by Apple and have common Normally try to avoid it. Plague – Because it reduces customer power and corporations love buyer power a lot.

Having said that, while Apple’s marketing efforts have declined significantly since the departure of Steve Jobs in both magnitude and execution, Apple still leads the PC market in marketing, indicating that it once was under Jobs. How much was ahead, even though the company was far less profitable. Then.

Marketing goes to the core of the GPU strategies that we will be discussing.

NVIDIA

Nvidia is arguably the segment leader and displays with it’s deepest line, and it’s the only one of the three with a Halo product. It’s interesting because on the CPU side, AMD’s Threadripper spun that market on its ear and allowed AMD to move aggressively on Intel’s workstation opportunity. It usurped key leadership positions for AMD and Lenovo, using it in part to gain segment dominance.

On the GPU side, the closest thing to AMD’s Threadripper is Nvidia’s 4090 RTX card, which is a beast. It performs more than most need and draws so much power that it’s melting power connectors and forcing power supply upgrades to run the part. But, in terms of raw performance, now no one in its segment can touch it.

A halo strategy uses an expensive showcase product to attract people to the brand and, if done correctly, can lead to increased sales of more affordable products that will sell in higher volumes. In the automobile market, it’s like having a Mercedes-Maybach or a supercar (Ford most recently used the Ford GT) to drive people to the dealership.

Since there is no such thing as an Nvidia store, the traffic aspect of the 4090 RTX remains untested and untapped. However, it still gives Nvidia performance bragging rights and solidifies its perceived market leadership, even though most buyers will be buying more affordable and underpowered alternatives.

amd

As mentioned, it’s interesting that, given AMD’s success with Threadripper, it hasn’t come up with a similar strategy, which would be a Threadripper CPU and Threadripper-like GPU in a killer workstation or gaming rig. For now, it’s the Nvidia 4090 RTX card.

Instead, AMD has pulled out two cards in the performance sweet spot. Unlike what Ford did with the Ford GT, GM did with the Corvette, specifically with the new C8. It provides most of the performance advantages that the Ford GT offers in a far more useful and affordable fashion, allowing the Corvette to massively outperform the Ford GT while costing a fraction of what the Ford costs.

The idea of ​​performance at a price you can afford resonates with buyers and doesn’t create some of the image problems that a halo product can create, such as excessive cost and excessive power use, resulting in negative coverage It is possible sustainability.

Buyers love the values, and while you don’t get the same performance bragging rights or visibility as with a Halo product, you also don’t have a problem with a part that’s pushing performance boundaries resulting in meltdowns. There may be problems like connectors (which Nvidia reports are still few in number and may be the result of assembly errors).

intel

Intel is new to this market and has two big problems. The first is that people know and trust GPU parts from AMD and Nvidia, and the second is that those two companies are chasing each other so hard that it’s hard for another firm to catch them from behind. Almost impossible. Intel needs to build market share and performance isn’t up to the top end, but it can deliver on features.

This approach is closer to how Honda entered the car market by creating an affordable part that offers features beyond its price class and where the real volume is, especially during a tight financial market where consumers are more willing to spend their money. are looking.

Right now, people are becoming very frugal, and Intel is working hard for the frugal buyer who only has the budget for an entry-level part. Intel’s cards aren’t performing competitively with Nvidia’s and AMD’s top offerings, but they’re a fraction of the price and have many of the same features typically not in cards in Intel’s sub-$300 price bracket.

Wrapping Up: The Best Strategy

The plan of action that best reflects the seller’s capabilities is the one to implement. The halo product strategy works for Nvidia as it must maintain its mark of the performance leader in the segment.

Given the success of AMD’s Threadripper, I’m surprised AMD hasn’t employed a similar strategy. Still, its approach to providing all the performance you need and creating an impression of superior performance value is in line with the historical brand image AMD has built against Intel, where it was historically the value play.

Intel doesn’t yet have the performance to compete with AMD or Nvidia at the high end, but it has a similar feature set, and the idea of ​​providing value play is the only way it’s open. The timing was fortunate for Intel as the market is looking for higher valuations given current economic conditions, providing a significant boost to Intel’s outlook.

So, Intel’s approach isn’t the only one the company can currently execute. Luckily, it was incredibly well timed, suggesting it will be far more successful this year than it otherwise would have been.

tech product of the week

surface pro 9

The Surface was built to compete with Apple and has many of the same features: a limited, focused line, a design-forward configuration that’s as much art as it is function, and a premium experience.

Microsoft missed out by making its stores Microsoft stores instead of Surface stores, so the stores tend to underperform their Apple counterparts. But Microsoft is a fundamentally different company and needs to keep Dell, Lenovo, HP and others happy, I understand how this decision came about. I think it was wrong, given its aim.

That said, the Surface Tablet was the primary weapon it had, when it first came to market, against the threat the iPad represented: a tablet that had long battery life, was extremely light, and at the time, effectively Was positioned as a laptop killer.

Steve Jobs, who piloted that product, died and since then Apple has decided not to replace the laptop with the iPad. Instead, it wants people to buy both, stalling the replacement effort.

But Microsoft didn’t get that memo. The initial surface was somewhat patchy. It would run Office, but its battery life was poor for a tablet. Over the years, both performance and battery life have improved.

This most recent Surface Pro 9 Arm Edition is the ultimate tablet. It does the best job of providing an iPad-like experience while still delivering PC-level performance. In fact, in some cases, it outperforms PC performance due to its unique AI NPU (Neural Processing Unit).

Microsoft Surface Pro 9

Surface Pro 9 | Image credit: Microsoft


I initially noticed a drop in performance in Microsoft Hello’s facial recognition capability. Normally, it takes a few seconds for a typical PC to recognize you, and this new Arm-based Surface Pro 9 does it almost instantly. My desktop PCs with specialized cameras are not as fast.

The video game I play most often is City of Heroes. On low performing PCs, it has to be scaled back to run on an arm machine, and it crashes or fails to load. However, on this new Surface Pro 9, it loaded and ran and even gave me advanced graphics options that I usually don’t get unless I’m running a discrete GPU (although, I hope , if I turn them on, the frame rate will be unplayable). Still, the fact that the game even loaded and was playable was a first with such a tablet, especially a running arm.

Another interesting aspect of this latest Surface Pro 9 tablet is that it accepts not only the proprietary magnetic Surface charger but also a generic USB-C charger. I really like the magnetic charger because I trip over the power cord, and if you pull on it the cable detaches and doesn’t cause the laptop to fall to the floor. On the other hand, a USB-C charger is much more useful as you can use a third-party charger, borrow a charger or even use it to charge your phone.

The optional pen and keyboard are now nicely nested together (I’m forever losing the pen). While I generally prefer a larger screen and a notebook form factor, this latest Surface was surprisingly usable and incredibly portable. At less than two pounds, with up to 19 hours of battery life, 5G WAN capability, and vastly improved video conferencing camera features (thanks to that NPU), I think the Surface Pro 9 is the future of ultra-portable products.

When Microsoft’s Surface Pro 9 launched, I knew it would be very different, and it is. It’s the first Arm-based product I’ve tested that not only feels compromised but outperforms its x86 counterpart in battery life, connectivity, and conferencing, which begs the question: What will the next generation do?

Ultimately, I think the Surface Pro 9 arm product is a milestone for the next generation of PCs, and it’s my product of the week. Let’s hope Microsoft does the same for the Surface Laptop!

The opinions expressed in this article are those of the author and do not necessarily reflect the views of ECT News Network.