The demands on server performance continue to increase at a tremendous pace. New requirements from large in-memory databases that are powering today’s cloud services and advanced analytics tools are arriving just as the impact of Moore’s Law is starting to slow. This is setting up a classic performance challenge that requires rethinking some of the core elements of today’s server architectures, particularly when it comes to memory. One key new opportunity is for high-speed server memory interface chipsets, which enable high-speed memory performance without compromising on memory capacities. Companies looking to optimize their server memory architecture designs, and improve their overall server performance and reliability, should give serious consideration to optimized DDR4 memory interface chipsets, which enhance the performance of server memory modules.
Papers
Protecting Pay TV with CryptoFirewall Security Cores
Download “Protecting Pay TV with CryptoFirewall Security Cores”
The advancing capabilities of attackers are driving a new phase in Pay TV signal piracy: pairing keys and control words can be extracted from set-top box (STB) chipsets. These attacks defeat the current defenses against control word sharing and provide access to plaintext control words, enabling illegal access to premium content by non-subscribers. Solving the problem requires improved security in device chipsets, or System-on-Chips (SoCs), including strong hardware security cores.Securing OTT Content with CryptoFirewall Security Cores
Download “Securing OTT Content with CryptoFirewall Security Cores”
Today, content protection in connected devices such as tablets and smart TVs is predominantly provided by software security. The level of protection achievable in software on these platforms today does not meet the latest studio requirements for premium content. In order to provide stronger security while simplifying product development, system-on-chips (SoCs) used in Internet-enabled televisions and mobile devices have started to include integrated hardware security cores. Once designed into an SoC, these hardware security cores provide a significant increase in security without sacrificing user-friendliness, flexibility or cost. The CryptoFirewall core is the leading hardware security core for TV/video distribution. It is designed to provide the multioperator flexibility and robust security required for Over-the-Top (OTT) multi-screen distribution to connected devices.Is Your Mobile Device Radiating Keys?
Is your mobile device’s EM emissions leaking your keys? A mobile app can inadvertently radiate secret data as cryptographic processing is done by the CPU. We’ll use a simple antenna and radio to perform live key extraction from several modern handheld devices. Developers can use several techniques to mitigate risk whenever applications use high-valued cryptographic keys.
Efficient sidechannel testing for public key algorithms: RSA case study
This paper proposes an approach to validate that implementations of public‐key cryptography have moderate resistance to side‐channel analysis, using RSA‐CRT as an example. The design goal of the proposed approach is to develop tests that are technically sound and repeatable, while at the same time being efficient and cost‐effective for testing labs. The approach was validated on two devices, one without countermeasures and another with some DPA countermeasures.
Download “Efficient side-channel testing for public key algorithms: RSA case study”
Intel Ivy Bridge Random Number Generator
Good cryptography requires good random numbers. This paper evaluates Intel’s hardware-based digital random number generator (RNG) for use in cryptographic applications.
Almost all cryptographic protocols require the generation and use of secret values that must be unknown to attackers. For example, random number generators are required to generate public/private keypairs for asymmetric (public key) algorithms including RSA, DSA, and Diffie-Hellman. Keys for symmetric and hybrid cryptosystems are also generated randomly. RNGs are used to create challenges, nonces (salts), padding bytes, and blinding values.
Because security protocols rely on the unpredictability of the keys they use, random number generators for cryptographic applications must meet stringent requirements. The most important property is that attackers, including those who know the RNG design, must not be able to make any useful predictions about the RNG outputs. In particular, the apparent entropy of the RNG output should be as close as possible to the bit length.
