AI chips: Explosive growth of deep learning is leading to rapid evolution of diverse, dedicated processors

/by

Artificial intelligence (AI) utilization has been accelerating rapidly for more than 10 years, as decreases in memory, storage and computation cost have made an increasing number of applications cost-effective. The technique of deep learning has emerged as the most useful. Large public websites such as Facebook (Nasdaq: FB) and Amazon (Nasdaq: AMZN), with enormous stores of data on user behavior and a clear benefit from influencing user behavior, were among the earliest adopters and continue to expand such techniques. Publicly visible applications include speech recognition, natural language processing and image recognition. Other high-value applications include network threat detection, credit fraud detection and pharmaceutical research.

Deep learning techniques are based on neural networks, inspired by animal brain structure. Neural networks perform successive computations on large amounts of data. Each iteration operates on the results of the prior computation, which is why the process is called “deep.” Deep learning relies on large amounts computation. In fact, deep learning techniques are well known; the recent growth is driven by decreasing costs of data acquisition, data transmission, data storage and computation. The new processors all aim to lower the cost of computation.

The new chips are less costly than CPUs for running deep learning workloads

Each computation is limited and tends to require relatively low precision, necessitating fewer bits than found in typical CPU operations. Deep learning computations are mostly tensor operations — predominantly matrix multiplication — and parallel tensor processing is the heart of many specialized AI chips. Traditional CPUs are relatively inefficient in carrying out this kind of processing. They cannot process many operations at the same time, and they deliver precision and capacity for complex computations that are not needed.

Nvidia (Nasdaq: NVDA) GPUs led the wave of new processors. In 2012, Google announced that its Google Brain deep learning project to recognize images of cats was powered by Nvidia GPUs, resulting in a hundredfold improvement in performance over conventional CPUs. With this kind of endorsement and with the widespread acceptance of the importance of deep learning, many companies, large and small, are following the money and investing in new types of processors. It is not certain that the GPU will be a long-term winner; successful applications of FPGAs and TPUs are plentiful.

Intel: Optimizing its scale advantage for Business of One flexibility

/by

TBR perspective

Usually sound business execution of world-class engineering, coupled with world-class monolithic manufacturing, has made Intel a dominant force around which technology businesses have orbited for decades. Intel’s dominance has been baked in the PC and server form factors, while ever smaller price points and form factors have shifted end-customer purchase criteria from computational performance specifications to business outcomes and user experiences.

Intel’s success has broadly expanded IT to address business problems and reshape our personal lives. Intel’s revenue growth prospects have diminished as its innovation has continued to increase the capacity and shrink the form factors and unit cost of its products. Intel delivers mature components that are embedded in mature products. Nevertheless, Intel thrives. The company has made mistakes, though, such as failing to address the mobile market. Intel’s capital- and engineering-intensive business requires it place large bets on its vision of the future. Now, facing waves of innovation in artificial intelligence (AI), Internet of Things (IoT) and processor design, Intel is, in effect, rearchitecting the company to reduce its dependence on the CPU, and thereby expand its market.

The key to Intel’s new architecture is companywide integration. Intel has always had more products and technologies, including video, networking, storage and memory silicon, than CPUs. As silicon becomes more diversified and is embedded in an increasing number of devices, Intel aims to create, along with customers, a far larger variety of solutions, often at a much smaller scale than the company’s monolithic products. To capitalize on the company’s enormous intellectual property, Intel must break down silos within the company. This will result in products that will often benefit from breaking down silos in silicon by facilitating the integration of computation, storage and communications.

The cultural challenge Intel will face will be in orchestrating and timing the various development teams such that the innovation cycles come together in world-class packages of tightly coupled compute, storage and networking form factors to power the smallest of edge compute instances and the largest of the high-performance computing (HPC) instances. The necessary work of rearchitecting the sales and marketing organizations remains for the next CEO, who has not yet been named, but the task is far less daunting than coordinating development and manufacture.

The thread that will stitch together these instances in the multicloud, always-on world of compute will be software. Software made interoperable through a “pruning,” as Intel Chief Engineering Officer and Technology, Systems Architecture & Client Group President Murthy Renduchintala described it, of the existing assets and frameworks into a cogent set of frameworks and tool sets to power innovation and optimize these scaled designs for specific workloads powered by AI is fed by voice and video as much as they have been fed by human interaction through keyboards in the past.

 

Intel Analyst Summit: Intel (Nasdaq: INTC) hosted an analyst event for the first time in four years to outline its technology road maps through 2021 and to articulate the business and cultural changes it believes are necessary for it to capitalize on the growing business opportunity Moore’s Law economics has unleashed. The senior leadership team gave about 50 analysts very detailed and frank briefings under a nondisclosure agreement (NDA), with ample time for follow-up conversations throughout the event.

CSPs accelerate NFV and SDN investments ahead of the 5G era

/by

Communication service providers (CSPs) are ramping up NFV- and SDN-related investments to realize greater cost savings and efficiencies, according to Technology Business Research Inc.’s (TBR) 2H18 Telecom Software Mediated Networks (NFV/SDN) Customer Adoption Study. This increase in investment will be driven by two underlying factors: CSPs under pressure to realize cost savings as their connectivity businesses remain under pressure and 5G pushing CSPs to pull forward their NFV and SDN road maps.

5G is greatly enhanced when using virtualization, especially when enabling and maximizing the benefits of network slicing and achieving better RAN economics. Though most CSPs intend to initially deploy the non-standalone (NSA) standard of 5G, which tethers 5G radio with EPC, an eventual upgrade to the standalone (SA) standard, which tethers 5G radio to a 5G core, will become a reality in the early 2020s. 5G core is inherently virtualized, and CSPs will be keen to prepare their networks to fully maximize the benefits of utilizing a virtualized network architecture, including, but not limited to, increasing agility, flexibility, visibility and cost efficiency.

 

 

TBR’s Telecom Software Mediated Networks (NFV/SDN) Customer Adoption Studyprovides an in-depth examination of how operators are planning, preparing and executing to succeed in the NFV and SDN market. TBR surveyed 50 people in operations, procurement and IT roles at 25 of the leading Tier 1 telecom service providers worldwide to gain insight into their NFV and SDN adoption plans. The study includes insight into service provider strategy, as well as service providers’ perceptions of supplier positioning and key benefits and obstacles.

5G-readiness spend and migration to new network architectures spur the TIS market to growth in 3Q18

/by

According to Technology Business Research, Inc.’s (TBR) 3Q18 Telecom Infrastructure Services (TIS) Benchmark, the TIS market grew as communication service provider (CSP) investment in areas tied to 5G-readiness increased. CSPs are rearchitecting their networks leveraging NFV, SDN and the cloud as well as implementing new business models, which requires growing spend across a broad range of professional services. Deployment services spend grew slightly, but the market will strengthen as the 5G spend cycle ramps up over the next couple of years, although the spend intensity will be lower than during the LTE cycle. RAN suppliers Nokia (NYSE: NOK), Ericsson, Huawei, ZTE and Samsung will capture incremental TIS market share as they drive high volumes of services attached to their 5G RAN. This is already occurring to some extent as CSPs densify networks as part of their 5G-readiness strategies. Though 5G will require significant hardware spend, the aggregate amount will be lower compared to LTE, which will drive vendors to explore new market areas, such as Industry 4.0.

The managed services market was flat year-to-year in 3Q18 as a decline in outsourcing was offset by growth in the out-tasking market. Generally, vendors are exercising pricing discipline when determining which outsourcing contracts to take on in an effort to improve margins. Ericsson is currently leading the way in this regard as it evaluates 42 contracts for exit or rescoping. Huawei, ZTE and CCS have been less concerned with price and are focused on consolidating the outsourcing market. Other vendors, including those that are historically hardware-centric with little to no footprint in the managed services market, are increasingly playing in out-tasking as they will manage applications deployed in CSP networks. Ciena (NYSE: CIEN) is an example of this trend.

 

 

TBR’s Telecom Infrastructure Services Benchmark provides quarterly analysis of the deployment, maintenance, professional services and managed services markets for network and IT suppliers. Suppliers covered include Accenture (NYSE: ACN), Amdocs, Atos, Capgemini, CGI, China Communications Services, Ciena, Cisco (Nasdaq: CSCO), CommScope, CSG International, Ericsson, Fujitsu, Hewlett Packard Enterprise (NYSE: HPE), Huawei, IBM (NYSE: IBM), Infosys (NYSE: INFY), Juniper Networks (NYSE: JNPR), NEC, Nokia (NYSE: NOK), Oracle (NYSE: ORCL), Samsung, SAP (NYSE: SAP), Tata Consultancy Services, Tech Mahindra, Wipro (NYSE: WIT) and ZTE.

U.S. 5G investment supports non-China-based vendors as Huawei and ZTE face increasing headwinds

/by

Nokia and Huawei are well-positioned to win as operators overhaul architectures in the 5G era, but most of the spend to date is on 5G radios, with Ericsson at an advantage due to market perception of its software-upgradeable Ericsson Radio System RAN. The network must ultimately be overhauled to fully realize 5G’s potential, but it will take CSPs many years to evolve their networks end-to-end, and the current focus — and 5G-related capex spend — will be on 5G radios. In the 5G RAN space, TBR believes Ericsson leads in market share. Nokia and Huawei, however, have broad portfolios that enable them to enter 5G accounts from multiple domains.

 

 

 

The Telecom Vendor Benchmark details and compares the initiatives and tracks the revenue and performance of the largest telecom vendors in segments including infrastructure, services and applications as well as in geographies including the Americas, EMEA and APAC. The report includes information on market leaders, vendor positioning, vendor market share, key deals, acquisitions, alliances, go-to-market strategies and personnel developments.

Ericsson’s turnaround is in process, but sustainability of business is in question

/by

TBR perspective

Though Ericsson’s focused strategy has proved to be a viable approach to stabilize the company, return it to profitability and provide incremental organic growth, the key concern will be how sustainable that stability and growth will be over the long term.

Ericsson’s focus on the wireless access domain tethers the company to the whims of that market, which is undergoing significant disruption as 5G and virtualization take hold and as operators increasingly shift capex budgets from connectivity infrastructure to building digital businesses, limiting Ericsson’s growth potential. Though there is room for Ericsson to take market share, particularly from Nokia (NYSE: NOK), Huawei and ZTE by leveraging its software-upgradable Ericsson Radio System (ERS) RAN gear, Ericsson is not immune to adverse business trends impacting the broader RAN market, namely legacy decommissioning, virtualization, openness, cloud and white box.

Ericsson is betting its ERS will offset the impact of these adverse trends and hasten its shift to a more software-centric entity with a more recurring, license-based software model that carries relatively high, sustainable margins, but this shift will take years to unfold and there is significant legacy business at risk of disappearing in the interim.

With the architecture of the network fundamentally changing to be virtualized and cloudified and communication service providers (CSPs) focused on relentless cost efficiency and TCO reduction, Ericsson will have to carefully balance its shift from the old world to the new reality, whereby forklift RAN upgrades become lower scale and targeted, and innovation and value migrate to the software layer. This has significant implications for Ericsson’s hardware and close-to-the-box services businesses, both of which are optimized to operate at high scale for efficiency and profitability.

TBR notes Ericsson and its close rival Nokia are pursing different paths during the 5G era. While Ericsson focuses on its core business of selling RAN and mobile core directly to service providers, Nokia is taking an end-to-end infrastructure approach and is building out a dedicated business unit with a full suite of resources to directly sell to enterprises. Though Industry 4.0, 5G and digital transformation are underlying themes that find commonality between the two vendors, their divergent tracks are noteworthy.

 

 

Ericsson (Nasdaq: ERIC) hosted its annual Industry Analyst Forum in Boston, bringing along a range of executives to provide an update on the company’s corporate and business unit strategies, with a focus on Networks, Managed Services and North America. Key topic areas included 5G, Internet of Things (IoT), automation and artificial intelligence (AI). Following the main session, analysts could attend three tracks — Network Evolution to 5G, AI and Automated Operations, or 5G and IoT Industry Innovation — and then participate in one-on-one speed meetings. The tone of Ericsson’s 2018 analyst day was upbeat as the company sees early signs that its turnaround plan is yielding results, evidenced by its 3Q18 earnings results in which organic revenue growth returned and margins improved markedly. Ericsson remains committed to its transformational restructuring and focused strategy, which are key pillars of its turnaround plan.

2019 Cloud & Software Predictions: More purchasing will be driven by intelligence rather than deployment

/by

In 2019 customers will care less about deployment, vendors will buy more, and both sides will become more intelligent

As 2018 comes to a close, it is good to look back at some of the long-standing questions that have been settled during the year:

  • Not all workloads will move to cloud; in fact, some will move back to on-premises delivery methods.
  • There will be no single cloud that rules all others, even if Amazon Web Services (AWS) remains the IaaS front-runner.
  • Wholesale migrations of legacy workloads will not occur quickly or easily.
  • The IT department will play an important role in cloud adoption moving forward.
  • Cloud needs to deliver short-term, immediate value for customers to consider adoption. The days of yearslong implementations are over.

The market is far from mature, however, as 2019 promises even more changes. Cloud remains a generic term for many customers, but they know just how many options are now available. A growing choice of delivery methods, combined with a much-improved management tool set, will drive pervasive hybrid adoption. These options allow customers to focus more on the solution, regardless of the specific cloud delivery method, that best fits their use case. Greater customer adoption is also creating more urgency among cloud providers, which will force more acquisition investments and riskier purchases as leaders scramble to best position themselves to claim share of the growing market. Lastly, the overarching trends around analytics, artificial intelligence (AI) and machine learning will become increasingly pervasive. Combined, these developments should redefine the cloud market yet again during 2019.

Opportunities for wireless subscriber growth remain plentiful for U.S. operators

/by

Cable providers are disrupting the U.S. wireless market

Subscriber growth for U.S. Tier 1 operators is being limited by the growing momentum of Comcast’s Xfinity Mobile brand, which outperformed AT&T and Sprint in postpaid phone net additions in 3Q18 and now has a base of over 1 million subscribers. Xfinity Mobile will become a stronger competitor in the U.S. market over the next several years as it expands its retail footprint and Comcast gains additional broadband customers to which it can cross-sell wireless service. Spectrum Mobile, which became available across Charter’s footprint in September, will also disrupt the U.S. wireless market by offering similar pricing incentives as Xfinity Mobile.

 

TBR’s U.S. & Canada Mobile Operator Benchmark details and compares the activities of the largest U.S. and Canadian operators, including financial performance, go-to-market initiatives and resource management strategies. Covered companies include AT&T (NYSE: T), Verizon (NYSE: VZ), Sprint (NYSE: S), T-Mobile (Nasdaq: TMUS), U.S. Cellular (NYSE: USM), Rogers, Telus and Bell Mobility.

2019 Services Predictions: Fix my business problem with a solution, not a slide — IT services and consulting for human-centric digital transformation

/by

Trust in turbulent times: data access and management as the key to IT services and consulting success in an uncertain 2019

From London to San Francisco, macroeconomic shifts and unsettled political environments in both the U.K. and the U.S. will make the start of 2019 turbulent and likely troubled for many companies, including the IT services vendors and consultancies we cover within TBR’s Professional Services practice. These companies will face harder decisions around repositioning their investments to other geographies or finding more cost-conscious investments in new areas. We are expecting a slowdown in both countries — not necessarily in revenues, but in fresh ideas and creativity, service launches, and expansion in additional markets within both economies — driven by new uncertainty and well-founded caution. The U.S. has traditionally been the largest market for IT services vendors, and no single year will change that hard, economic fact. But where we have seen IT services vendors experiment with new consulting business models that blend emerging technologies into strategy consulting and embed codeveloped IP into outcomes-based IT services engagements, we expect a retrenchment as 2019 opens, with uncertainty lingering at least through the summer. By this time next year, we expect to see more initiatives in APAC leveraging that region’s faster adoption of 5G (and all that means for digital transformation at speed and scale). And we expect the three trends described below will be demonstrably evident in the strategies and performances of the leading IT services vendors and consultancies we cover.

Underlying all of our assessments, we are developing a new appreciation for the criticality of data. Beyond the cliché that every company is a data company or that data is the new oil, we have been seeing throughout 2018 the way IT services vendors and consultancies have begun investing increasingly in data management, cleansing and protection, all with the assumption that analytics, automation, artificial intelligence and every other emerging technology starts with and relies on clean and useful data. Smarter business decisions do not come from bad data, no matter how good the algorithm or analytics package. For 2019, this means data access becomes an opportunity to extend to all IT services, up to and including digital transformation, the same trust that comes with an audit or a multiyear outsourcing engagement. Imagine a consultancy working with unfettered access to every data element across a client’s enterprise. Getting there may take a changed regulatory environment and will definitely require that boards be willing to extend trust in new ways, a human limitation that may slow this new data access. But we see it coming. If politics and economics could cause stormy weather in the U.S. and U.K., the acceleration of digital transformation through data access may be the longer-term trend, the global warming lifting all boats on rising sea levels.

Don’t stop thinking about tomorrow: Amazon, RPA, AI and ethical IT in the federal sector

/by

Notwithstanding the increased integration of artificial intelligence (AI) and process bots into government operations, the U.S. federal services sector decidedly remains a people business. At a recent Washington Technology Power Breakfast forum, industry leaders talked talent strategies and how they hope to succeed as digital transformation fundamentally changes the types of people sought for government work. A few key themes emerged as near-universal top-of-mind concerns for forum participants and audience members, such as the importance of developing a brand and messaging values that resonate with the emerging workforce; the criticality of public-private partnerships to develop talent in the greater Washington, D.C., area and beyond; and the concern and uncertainty about the human capital impact of Amazon’s (Nasdaq: AMZN) recent decision to become a much closer neighbor of Uncle Sam.

The trends and issues discussed often repeated themes TBR touches on regularly in its analysis of the IT industry, both within the federal market and across public and private sectors globally. While the perspectives shared were both validating and enlightening, there was just as much value in paying attention to what the panelists did not talk about at length. Today’s pressing HR demands leave little time for talent strategists to worry about the looming disruptive impacts of AI and robotic process automation (RPA), the fundamental changes in labor amid the rise of asset-based services, forward-thinking venture-capital-like approaches to partnerships, or the uncomfortable and growing issue of ethics conflicting with the eagerness to apply innovative IT to government missions. HR leaders and strategic decision makers at the leading services firms will need to grapple with these difficult topics today if they want to stay ahead of disruption that is just around the corner in the dynamic and rapidly changing IT industry.

 

 

Washington Technology Power Breakfast: TBR Public Sector Analyst Joey Cresta was recently invited to participate in a panel discussion on talent strategies of government contractors at a breakfast forum hosted by Washington Technology. The event provided an outlet for executives, HR experts and industry thought leaders to share how they intend to win talent in a competitive labor market while maintaining profitability and bracing for the impact of Amazon’s impending move into Crystal City.