Highlights:

• Intel Corporation announced two lines of processors meant to power workstations, which are high-end Windows computers used for simulations and data research software.
• The W9-3495X is the fastest processor introduced by Intel recently. It contains up to 56 cores with a maximum clock speed of 4.8 gigahertz.

Recently, Intel Corporation announced two lines of processors meant to power workstations. These are high-end Windows computers used for simulations and data research software.

The Xeon W-2400 series is the first processor series from the recent launch. It consists of eight chips intended to power ordinary workstations. Intel is also releasing the more powerful Xeon W-3400 series, composed of seven chips designed for high-end workstations.

Roger Chandler, vice president and general manager of creator and workstation solutions at Intel, said, “Our new Intel Xeon desktop workstation platform is uniquely designed to unleash the innovation and creativity of professional creators, artists, engineers, designers, data scientists, and power users – built to tackle both today’s most demanding workloads as well as the professional workloads of the future.”

The W9-3495X is the fastest processor introduced by Intel. It contains up to 56 cores with a maximum clock speed of 4.8 gigahertz. Compared to Intel’s previous workstation silicon, the W9-3495X performs single-thread workloads up to 28% faster and multithread apps 120% faster.

The W9-3495X consists of four chiplets or computation modules with several cores apiece. Using a method called EMIB, Intel makes the chiplets independently and then combine them into a single CPU. EMIB helps Intel to manufacture CPUs at a lower cost than competing technology on the market.

The chiplets in a central processing unit must be able to share data to perform computations. Typically, manufacturers meet this need by mounting chips on an interposer, a rectangular piece of silicon. It comprises wires via which data may be exchanged between the various modules of a CPU.

Interposers allow rapid data transmission rates, which enhances the performance of processors. However, they may be difficult and expensive to produce. Intel’s new W9-3495X flagship workstation processor incorporates EMIB technology to alleviate this restriction.

EMIB features a small interposer that is easier to fabricate than conventional versions of the technology. According to Intel, the consequence is decreased production complexity and expense.

The four chiplets that make up the Intel W9-3495X chip are positioned atop an EMIB-powered base layer. The base layer incorporates a tiny interposer that enables data transfer between chiplets. It comprises the cables that deliver power from the motherboard to the CPU.

Intel’s EMIB technology, which was purportedly unavailable in the company’s older workstation CPUs, is only one factor in the W9- 3495X’s performance. Intel has also made a variety of other enhancements to the chip.

The W9-3495X has a 105-megabyte L3 cache that stores data near the chip’s logic circuitry, reducing data travel times. The outcome is a performance improvement. The chip also has Intel Turbo Boost Max Technology 3.0, which directs demanding programs to the processor’s fastest cores.

Several of the other processors introduced by Intel recently also include EBIM technology. Others support overclocking, which boosts a processor’s frequency over its default maximum speed. Additionally, some of the chips may overclock the RAM of the machine in which they are placed, enhancing performance even further.

The seven processors of the W-3400 series, the faster of the two processor lineups for workstations introduced by Intel recently, include between 12 and 56 cores. In contrast, the W-2400 series consists of eight chips with six to 24 cores.

Many workstations have graphics and central processing units to accelerate rendering and machine-learning applications. Intel chips are typically paired with Nvidia Corporation GPUs. Nvidia recently stated that several computer manufacturers want to introduce workstations that use Intel’s newly announced CPUs with its RTX 6000 graphics card.

The RTX 6000 contains 18,176 CUDA cores, which function similarly to CPU processing cores. The chip also has circuits tuned for running ray tracing rendering algorithms. Tensor Cores, a third class of specialized circuits, can accelerate artificial intelligence applications.

Nvidia asserts that specific workstations can integrate the RTX 6000 with its ConnectX-6 Dx SmartNIC chip. The chip is utilized for networking functions, including data transfer between computers. According to the corporation, workstation users could download and exchange complicated datasets faster.