THE CHALLENGE

In hyperscale data centers, the most expensive component in CPU-based servers, in terms

THE CHALLENGE

In hyperscale data centers, the most expensive component in CPU-based servers, in terms of both monetary cost and carbon footprint, is currently computer memory; going forward, memory is expected to become even more expensive relative to other server components. Hardware memory compression, where the CPU’s memory controller transparently compresses and packs memory values more densely, is a promising and performant solution to combat the high cost of memory. While this approach boosts memory capacity without adding expensive memory chips, it also breaks the traditional link between what the operating system (OS) thinks it is using and what the hardware actually stores, since dynamically-compressed data takes up varying amounts of real space. As a result, the OS can no longer accurately partition memory resources across different co-located workloads, leading to increased performance variation. Current tools like memory quotas and ballooning lack visibility into this compressed layer, making it hard to guarantee performance or isolate workloads in multi-tenant environments, ultimately affecting server efficiency, cost savings, and the predictability that enterprises and cloud customers’ demand.

OUR SOLUTION

We introduce a hardware innovation called Multi-domain Hardware Memory Compression to embed directly into each CPU’s memory controller to give cloud providers and data center operators fine-grained control over how much actual DRAM each workload consumes even when memory is compressed in hardware. By letting the OS assign clear memory quotas through lightweight per-job control blocks, Multi-domain Hardware Memory Compression ensures that each job stays within its real memory budget by automatically compressing less-used data in the background. If a workload still exceeds its quota, the system is alerted early to take corrective action. Unlike previous methods that relied on software guesses about data compressibility, this solution enforces exact memory limits in hardware, delivering predictable performance and strong isolation between tenants. The result is tighter consolidation, reduced overprovisioning, and up to 70% savings in memory costs while maintaining the reliability and service-level guarantees that businesses and cloud customers require.

Figure: Full-system prototype on a Genesys 2 Kintex-7™ FPGA. Linux boots up with two cores and 3979736 KB – 4X the 1GB DRAM on board (prototype: https://youtu.be/-1JG3JnIY3U).

Advantages:

• Precise per-job machine-physical memory quotas enforced in hardware
• Stable multi-tenant performance with very low variability under colocation
• Objective-driven, low-latency compression without OS compressibility estimates
• Seamless OS integration enabling higher consolidation and memory cost savings

Potential Application:

• Cloud server memory consolidation
• Multi-tenant performance isolation
• High-performance computing optimization
• Container memory quota enforcement

This innovation repurposes an FDA-approved antibody to treat and prevent aortic stiffening and resistant hypertension

This innovation repurposes an FDA-approved antibody to treat and prevent aortic stiffening and resistant hypertension (RH). This mechanism-based therapeutic, inhibits receptor mediated inflammatory and pro-stiffness signaling to reduce aortic stiffness and improve blood pressure control, ultimately breaking the root cycle of hypertension.

Background: 
Hypertension is the leading cause of cardiovascular disease (CVD) and premature mortality, with prevalence increasing with age. Despite the availability of multiple medications, more than half of patients’ blood pressure (BP) remains uncontrolled, and millions develop resistant hypertension, which is defined as persistently elevated BP despite treatment with three or more antihypertensive drug classes, including a diuretic. RH affects over 10 million individuals in the U.S. and approximately 200 million worldwide, contributing more than $1 trillion annually to U.S. healthcare costs, with its burden rising as the population ages.

Persistent uncontrolled hypertension leads to life-threatening complications such as stroke, heart failure, aortic dissection, renal failure, and premature death. The incidence of RH rises sharply with age, largely driven by aortic stiffening, a process accelerated by both aging and high BP, creating a vicious cycle of vascular dysfunction. Current therapies primarily target hemodynamic factors (e.g., vasodilation or sodium balance) but do not to reverse aortic stiffness, leaving a critical gap in the management of RH.

Applications: 

• Resistant hypertension therapeutic 
• Aortic stiffness therapeutic 
• Immunological therapeutic 
• Cardiovascular disease drug development

Advantages: 

• Repurposes an FDA-approved drug for novel use
• Minimizes development risk and cost in a shortened time
• Biological targeting of inflammation-driven aortic stiffening in RH
• Improve blood pressure control
• Potential to expand therapeutic applications to broader cardiovascular complications such as stroke, heart failure, aortic dissection, renal diseases
• Preventative approach to treat RH and aortic stiffening

This technology introduces a novel separation method designed to isolate small protein biomarkers directly from saliva

This technology introduces a novel separation method designed to isolate small protein biomarkers directly from saliva, enabling a rapid and non-invasive diagnostic approach for cardiac infarction. The system employs a hybrid chromatographic matrix that selectively rejects large proteins while allowing smaller biomarkers to interact with affinity sites and be effectively recovered. This streamlined process provides the foundation for saliva-based test kits that can support early detection and monitoring of heart damage and other diseases. The innovation not only broadens access to simple diagnostic tools but also reduces reliance on invasive blood sampling and lengthy lab work.

Background: 
Cardiac infarction remains one of the leading causes of mortality worldwide, and timely detection is critical to improving patient outcomes. Current diagnostic practices typically require blood draws, which can be invasive, time-consuming, and resource intensive. Saliva represents a promising alternative because of its ease of collection, but existing methods have struggled to isolate low-molecular-weight biomarkers due to interference from larger proteins. Traditional separation techniques often require multiple complex steps, making them costly and inefficient. This invention directly addresses these shortcomings by integrating size exclusion with affinity chromatography in a single platform, enabling precise recovery of disease-linked biomarkers from saliva in a faster and simpler manner than conventional approaches.

Applications: 

• Cardiac infarction detection and monitoring
• Broader cardiovascular disease diagnostics
• Development of saliva-based diagnostic kits for clinical or home use
• Expansion into oncology and other diseases with salivary biomarkers

Advantages: 

• Enables non-invasive, saliva-based diagnostics
• Provides earlier and more accessible detection of cardiac infarction
• Reduces time and cost compared to blood-based methods
• Streamlines sample processing in one system
• High selectivity for low-molecular-weight biomarkers, minimizing sample interference
• Potential for point-of-care and at-home test kit development

This technology is a scanning thermal microscope device. It utilizes a thermocouple with a sharp tip on its

This technology is a scanning thermal microscope device. It utilizes a thermocouple with a sharp tip on its center fabricated from a cross-junction of coated and uncoated thin films and is suspended on a nanowire device. Radiation heat received at the tip is converted into a Seeback voltage signal for thermoelectric temperature sensing. This device can operate without making physical contact with the object being scanned, enabling minimally invasive sensing, as well as with minimal direct contact between the tip and the sample. It delivers exceptionally high temperature sesnsing, offering two orders of magnitude of improvement in temperature accuracy over an Au-Ni thermocouple. 

Background: 
Electronic components are becoming smaller at an exponential rate, and the demand for high-performing, increasingly small electronics continues to grow in a wide range of industries. However, overheating is a major problem in microelectronics that has become a bottleneck to performance, reliability, and lifespan. There’s a need for technologies to sense temperatures across microelectronic devices so that points of increased overheating risk can be identified and changes can be made to improve thermal management. Existing temperature mapping techniques have many limitations. Many require contact with the device, which can potentially cause damage, and even non-contact solutions can damage electronic components due to temperature differences between the thermocouple tip and the scanned surface and heat conduction through the air and water meniscus between the tip and the sample. This technology provides a non-contact, silicon-based thermocouple for minimally invasive, highly accurate thermal microscopy. 

Applications: 

• Electronics
• Semiconductors
• Nanoscale temperature mapping
• Thermometry
• Device diagnostics

Advantages: 

• Increased accuracy and sensitivity in temperature readings
• Supports both direct contact and non-contact measurements
• Improves thermal management at the nanoscale level