Spectre and Meltdown, two methods of exploiting security vulnerabilities found in Intel, AMD and Arm processors, demonstrate weaknesses in current hardware cybersecurity that will force a huge paradigm shift within the semiconductor industry.
Software-based cybersecurity, the go-to measure to ensure a system won’t be hacked, addresses software vulnerabilities but overlooks hardware design. That’s because more than $150 billion is spent a year on software-based cyber security tools, while relatively little is spent on hardware security tools, and there continues to be a stream of hacks and breaches.
As machines control more of our physical world, security needs to be built in from the ground up, utilising the latest security technologies to protect software and hardware.
The gap between the intent of security IP building blocks and their actual deployment in full SoC designs must be filled. What’s needed is a proactive and early approach to identifying and eliminating security vulnerabilities throughout the design of a semi-conductor chip. While a software vulnerability can often be patched, a hardware vulnerability in silicon deployed in systems is very costly to repair.
Ensuring the chip’s final implementation does not expose a security hole that software will exploit is a clear call to action. Without these solutions, chips will continue to be built in ways that leave them vulnerable to hackers.
The chip verification investment today is driven by requirements of functional verification. While absolutely essential, it is this focus on functionality that can lead to the introduction of unintentional security vulnerabilities during the design and verification cycle.
IoT designs may be the most vulnerable and Smart IoT devices will push the edge further from the enterprise expanding the size of the core network. Their volume will increase by 10-to-100 fold as this segment continues to accelerate. Huge investments in the end-to-end ecosystem will support this expansion.
However, unless investments in hardware security increase significantly as the interconnectedness expands, the risk and liability to both service providers as well as the edge consumer will increase.
The methodology and techniques to verify hardware security must catch up to the complexity of the SoCs that implement them. Fortunately, the shift from conversation to action is beginning as silicon providers feel the impact of gaps in security exposed in deployed products.
Investment in development of secure silicon architectures and foundation building blocks has been increasing for some time. Investment for hardware security is now increasing as well because hardware threat scenarios must be verified before products are released and deployed in the communications infrastructure.
As a result, chip design is moving from a focus on verification of functionality to verification for security.
This paradigm shift will create new de-facto standards and methodologies that must be deployable without increasing the overall SoC verification schedule. Ideally, they will co-exist with existing verification processes that yield an overall reduction in project schedule, with a significant reduction in security vulnerability.
Only then will they be adopted as standard practice in time-sensitive projects servicing the compute and mobile communications market, and the safety critical markets of automotive and aerospace.
Source: Cyber Security Intelligence
So what have you done lately? Working on preventative measures? Assemble and train the ‘crisis team’? Have regular ‘dry-runs’ by the crisis team?