On the dangers of Intel’s frequency scaling

While I was writing the post comparing the new Qualcomm server chip, Centriq, to our current stock of Intel Skylake-based Xeons, I noticed a disturbing phenomena.

When benchmarking OpenSSL 1.1.1dev, I discovered that the performance of the cipher ChaCha20-Poly1305 does not scale very well. On a single thread, it performed at the speed of approximately 2.89GB/s, whereas on 24 cores, and 48 threads it performed at just over 35 GB/s.

CC BY-SA 2.0 image by blumblaum

Now this is a very high number, but I would like to see something closer to 69GB/s. 35GB/s is just 1.46GB/s/core, or roughly 50% of the single core performance. AES-GCM scales much better, to 80% of single core performance, which is understandable, because the CPU can sustain higher frequency turbo on a single core, but not all cores.

alt

Why is the scaling of ChaCha20-Poly1305 so poor? Meet AVX-512. AVX-512 is a new Intel instruction set that adds many new 512-bit wide SIMD instructions and promotes most of the existing ones to 512-bit. The problem with such wide instructions is that they consume power. A lot of power. Imagine a single instruction that does the work of 64 regular Continue reading

Rough Guide to IETF 100: Identity, Privacy, and Encryption

Identity, privacy, and encryption continue to be active topics for the Internet Society and the IETF community impacting a broad range of applications. In this Rough Guide to IETF 100 post, I highlight a few of the many relevant activities happening next week in Singapore, but there is much more going on so be sure to check out the full agenda online.

Encryption

Encryption continues to be a priority of the IETF as well as the security community at large. Related to encryption, there is the TLS working group developing the core specifications, several working groups addressing how to apply the work of the TLS working group to various applications, and the Crypto-Forum Research Group focusing on the details of the underlying cryptographic algorithms.

The Transport Layer Security (TLS) working group is a key IETF effort developing core security protocols for the Internet. This week’s agenda includes both TLS 1.3 and Datagram Transport Layer Security. Additionally, the TLS working group will be discussing connection ID, exported authenticators, protecting against denial of service attacks, and application layer TLS. The TLS working group is very active and, as with all things that are really important, there are many Continue reading

Deploy360’s Hot Topics at IETF 100

Next week is IETF 100 in Singapore which will be the first time the IETF has been held in the country. The Deploy360 team will be represented by Megan Kruse and Dan York, along with ISOC’s Chief Internet Technology Officer Olaf Kolkman. We’re again highlighting the latest IPv6, DNSSEC, Securing BGP, TLS and IoT related developments.

Below are the sessions that we’ll be following. Note this post was written in advance so please check the official IETF 100 agenda for any updates, room changes, or final details.

Monday, 13 November 2017

Tuesday, 14 November 2017

Qualcomm ships ARM chip to challenge Intel in cloud data centers

Qualcomm's much-anticipated ARM-based Centriq 2400 product line, which started shipping commercially this week, is a worthy contender to break Intel's virtual monopoly in the server processor arena, where data center operators are thirsting to see competition to help bring down costs.An unsolicited acquisition bid for Qualcomm from Broadcom, emerging server-chip competitors and legal wrangles involving Apple and other vendors, however, cast a bit of a shadow on prospects for the new chip.[ Check out this review: How rack servers from HPE, Dell and IBM stack up . ]To read this article in full, please click here

Qualcomm ships ARM chip to challenge Intel in cloud data centers

Qualcomm's much-anticipated ARM-based Centriq 2400 product line, which started shipping commercially this week, is a worthy contender to break Intel's virtual monopoly in the server processor arena, where data center operators are thirsting to see competition to help bring down costs.An unsolicited acquisition bid for Qualcomm from Broadcom, emerging server-chip competitors and legal wrangles involving Apple and other vendors, however, cast a bit of a shadow on prospects for the new chip.[ Check out this review: How rack servers from HPE, Dell and IBM stack up . ]To read this article in full, please click here

Qualcomm ships ARM chip to challenge Intel in cloud data centers

Qualcomm's much-anticipated ARM-based Centriq 2400 product line, which started shipping commercially this week, is a worthy contender to break Intel's virtual monopoly in the server processor arena, where data center operators are thirsting to see competition to help bring down costs.An unsolicited acquisition bid for Qualcomm from Broadcom, emerging server-chip competitors and legal wrangles involving Apple and other vendors, however, cast a bit of a shadow on prospects for the new chip.Qualcomm revealed some impressive specs at an industry event in San Diego Wednesday, bringing out a a variety of big-time cloud, hardware and software providers to show support.To read this article in full, please click here

Qualcomm ships ARM chip to challenge Intel in cloud data centers

Qualcomm's much-anticipated ARM-based Centriq 2400 product line, which started shipping commercially this week, is a worthy contender to break Intel's virtual monopoly in the server processor arena, where data center operators are thirsting to see competition to help bring down costs.An unsolicited acquisition bid for Qualcomm from Broadcom, emerging server-chip competitors and legal wrangles involving Apple and other vendors, however, cast a bit of a shadow on prospects for the new chip.Qualcomm revealed some impressive specs at an industry event in San Diego Wednesday, bringing out a a variety of big-time cloud, hardware and software providers to show support.To read this article in full, please click here

HPE Developing its Own Low Power “Neural Network” Chips

With so many chip startups targeting the future of deep learning training and inference, one might expect it would be far easier for tech giant Hewlett Packard Enterprise to buy versus build. However, when it comes to select applications at the extreme edge (for space missions in particular), nothing in the ecosystem fits the bill.

In the context of a broader discussion about the company’s Extreme Edge program focused on space-bound systems, HPE’s Dr. Tom Bradicich, VP and GM of Servers, Converged Edge, and IoT systems, described a future chip that would be ideally suited for high performance computing under …

HPE Developing its Own Low Power “Neural Network” Chips was written by Nicole Hemsoth at The Next Platform.

IDG Contributor Network: How blockchain technology could affect the future of network engineering

Blockchain, also referred to as distributed ledger, is the concept behind the success of Bitcoin and provides a dynamic digital register of transactions. Think of it as a database that’s distributed throughout a network. Information is continually shared and reconciled throughout multiple nodes and each one has an identical copy of the database. Transactions within this database are audited and agreed upon by consensus. This decentralized method of keeping track of changes ensures the ledger can’t be practically controlled by any one entity, eliminates the possibility of single-points of failure, and allows for the verification of transactions without the need for third-party intervention. Since each interaction is public, blockchain technology offers a reliable, incorruptible transaction-based infrastructure and the value it provides isn’t just limited to cryptocurrency.To read this article in full, please click here

IDG Contributor Network: How blockchain technology could affect the future of network engineering

Blockchain, also referred to as distributed ledger, is the concept behind the success of Bitcoin and provides a dynamic digital register of transactions. Think of it as a database that’s distributed throughout a network. Information is continually shared and reconciled throughout multiple nodes and each one has an identical copy of the database. Transactions within this database are audited and agreed upon by consensus. This decentralized method of keeping track of changes ensures the ledger can’t be practically controlled by any one entity, eliminates the possibility of single-points of failure, and allows for the verification of transactions without the need for third-party intervention. Since each interaction is public, blockchain technology offers a reliable, incorruptible transaction-based infrastructure and the value it provides isn’t just limited to cryptocurrency.To read this article in full, please click here

HPC Heavyweight Goes All-In On OpenACC

Across the HPC community, commercial firms, government labs and academic institutions are adapting their code to embrace GPU architectures. They are motivated by the faster performance and lower energy consumption provided by GPUs, and many of them are using OpenACC to annotate their code and make it GPU-friendly. The Next Platform recently interviewed one key organization to learn why it is using the OpenACC programming model to expand its computing capabilities and platform support.

If the earth was the size of a basketball, its atmosphere would be the thickness of shrink wrap. It is fragile enough that in 1960, the …

HPC Heavyweight Goes All-In On OpenACC was written by Timothy Prickett Morgan at The Next Platform.

1 … 1,821 1,822 1,823 1,824 1,825 … 3,833