The proliferation of Large Language Models (LLMs) in diverse applications underscores the pressing need for robust security measures to thwart potential jailbreak attacks. These attacks exploit vulnerabilities within LLMs, endanger data integrity and user privacy. Guardrails serve as crucial protective mechanisms against such threats, but existing models often fall short in terms of both detection accuracy, and computational efficiency. This paper advocates for the significance of jailbreak attack prevention on LLMs, and emphasises the role of input guardrails in safeguarding these models. We introduce MoJE (Mixture of Jailbreak Expert), a novel guardrail architecture designed to surpass current limitations in existing state-of-the-art guardrails. By employing simple linguistic statistical techniques, MoJE excels in detecting jailbreak attacks while maintaining minimal computational overhead during model inference. Through rigorous experimentation, MoJE demonstrates superior performance capable of detecting 90% of the attacks without compromising benign prompts, enhancing LLMs security against jailbreak attacks.

The evolving complexity of modern IT infrastructures has paved the way for malicious actors to exploit a wide array of vulnerabilities that can compromise the integrity of these systems. Monitoring complex IT systems is expensive and often requires dedicated infrastructure for deploying Intrusion and/or Anomaly Detection Systems. Moreover, ML-based solutions need large training sets, which add to the overall cost. To tackle these challenges we present INTELLECT, a novel approach to Intrusion and/or Anomaly Detection System, which leverages Federated Learning and model pruning techniques to cooperatively train high-accuracy models using distributed datasets and derive a fleet of lightweight models, which can be deployed without incurring additional costs for dedicated infrastructure. INTELLECT expands on the state-of-the-art techniques for feature selection, model pruning, and model distillation to create an interconnected pipeline. We empirically demonstrate the effectiveness of the methodology on benchmark datasets, and we present guidelines for the deployment in production systems.

In this paper, we present a new scalable and adaptive architecture for FL aggregation. First, we demonstrate how traditional tree overlay based aggregation techniques (from P2P, publish-subscribe and stream processing research) can help FL aggregation scale, but are ineffective from a resource utilization and cost standpoint. Next, we present the design and implementation of AdaFed, which uses serverless/cloud functions to adaptively scale aggregation in a resource efficient and fault tolerant manner. We describe how AdaFed enables FL aggregation to be dynamically deployed only when necessary, elastically scaled to handle participant joins/leaves and is fault tolerant with minimal effort required on the (aggregation) programmer side. We also demonstrate that our prototype based on Ray [1] scales to thousands of participants, and is able to achieve a > 90% reduction in resource requirements and cost, with minimal impact on aggregation latency.

With the lowering costs of sensors, high-volume and high-velocity data are increasingly being generated and analyzed, especially in IoT domains like energy and smart homes. Consequently, applications that require accurate short-term forecasts and predictions are also steadily increasing. In this paper, we provide an overview of a novel end-to-end platform that provides efficient ingestion, compression, transfer, query processing, and machine learning-based analytics for high-frequency and high-volume time series from IoT. The performance of the platform is evaluated using real-world dataset from RES installations. The results show the importance of high-frequency analytics and the surprisingly positive impact of error bounded lossy compression on machine learning in the form of AutoML. For example, when detecting yaw misalignments in wind turbines, an improvement of 9% in accuracy was observed for AutoML models on lossy compressed data compared to the current industry standard of 10-minute aggregated data. Thus, these small-scale experiments show the potential of the platform, and larger pilots are planned.

The transition away from carbon-based energy sources poses several challenges for the operation of electricity distribution systems. Increasing shares of distributed energy resources (e.g. renewable energy generators, electric vehicles) and internet-connected sensing and control devices (e.g. smart heating and cooling) require new tools to support accurate, data-driven decision making. Modelling the effect of such growing complexity in the electrical grid is possible in principle using state-of-the-art power-power flow models. In practice, the detailed information needed for these physical simulations may be unknown or prohibitively expensive to obtain. Hence, data-driven approaches to power systems modelling, including feed-forward neural networks and auto-encoders, have been studied to leverage the increasing availability of sensor data, but have seen limited practical adoption due to lack of transparency and inefficiencies on large-scale problems. Our work addresses this gap by proposing a data- and knowledge-driven probabilistic graphical model for energy systems based on the framework of graph neural networks (GNNs). The model can explicitly factor in domain knowledge, in the form of grid topology or physics constraints, thus resulting in sparser architectures and much smaller parameters dimensionality when compared with traditional machine-learning models with similar accuracy. Results obtained from a real-world smart-grid demonstration project show how the GNN was used to inform grid congestion predictions and market bidding services for a distribution system operator participating in an energy flexibility market.

Abstract Electricity is a major cost in running a data centre, and servers are responsible for a significant percentage of the power consumption. Given the widespread use of HTTP, both as a service and a component of other services, it is worthwhile reducing the power consumption of web servers. In this paper we consider how reverse proxies, commonly used to improve the performance of web servers, might be used to improve energy efficiency. We suggest that when demand on a server is low, it may be possible to switch off servers. In their absence, an embedded system with a small energy footprint could act as a reverse proxy serving commonly-requested content. When new content is required, the reverse proxy can power on the servers to meet this new load. Our results indicate that even with a modest server, we can get a 25% power saving while maintaining acceptable performance.

For most multi-threaded applications, data structures must be shared between threads. Ensuring thread safety on these data structures incurs overhead in the form of locking and other synchronization mechanisms. Where data is shared among multiple threads these costs are unavoidable. However, a common access pattern is that data is accessed primarily by one dominant thread, and only very rarely by the other, non-dominant threads. Previous research has proposed biased locks, which are optimized for a single dominant thread, at the cost of greater overheads for non-dominant threads. In this article we propose a new family of biased synchronization mechanisms that, using a modified interface, push accesses to shared data from the non-dominant threads to the dominant one, via a novel set of message passing mechanisms. We present mechanisms for protecting critical sections, for queueing work, for caching shared data in registers where it is safe to do so, and for asynchronous critical section accesses. We present results for the conventional Intel® Sandy Bridge processor and for the emerging network-optimized many-core IBM® PowerEN™ processor. We find that our algorithms compete well with existing biased locking algorithms, and, in particular, perform better than existing algorithms as accesses from non-dominant threads increase.

Understanding the baseline underwater acoustic signature of an offshore location is a necessary, early step in formulating an environmental impact assessment of wave energy conversion devices. But in order to even begin this understanding, infrastructure must be deployed to capture raw acoustic signals for an extended period of time. This infrastructure is comprised of at least four distinct components. Firstly, a hydrophone, deployed underwater, which is capable of operating at a high sampling rate: 500,000 16-bit samples per second. Secondly, an analog/digital converter (ADC), to which the hydrophone transmits raw voltages. Thirdly, a communications infrastructure for bridging the gap from the ADC to shore. And finally, an onshore base-station for receiving the signals and presenting them to a remote analytic or simulation infrastructure for further processing. Attempting this signal capture in real-time poses many problems. On a practical level, deploying cabled infrastructure to deliver power and communications to the offshore components may be prohibitively expensive. However, reliance on solar power may result in interruptions to real-time wireless transmission. Additionally, a high sampling rate will require significant base-station memory/storage/processing capabilities as well as potentially high costs of delivery to a remote infrastructure, part of which could be alleviated by real-time signal compression. This paper discusses our attempts at implementing such a system which would reliably acquire real-time data and scale with growing demands.

SUMMARY Unconventional multi‐ and many‐core processors (e.g. IBM ® Cell B.E. TM and NVIDIA ® GPU) have emerged as effective accelerators in trial climate and weather simulations. Yet these climate and weather models typically run on parallel computers with conventional processors (e.g. Intel ®, AMD ®, and IBM) using Message Passing Interface. To address challenges involved in efficiently and easily connecting accelerators to parallel computers, we investigated using IBM's Dynamic Application Virtualization TM (IBM DAV) software in a prototype hybrid computing system with representative climate and weather model components. The hybrid system comprises two Intel blades and two IBM QS22 Cell B.E. blades, connected with both InfiniBand ® (IB) and 1‐Gigabit Ethernet. The system significantly accelerates a solar radiation model component by offloading compute‐intensive calculations to the Cell blades. Systematic tests show that IBM DAV can seamlessly offload compute‐intensive calculations from Intel blades to Cell B.E. blades in a scalable, load‐balanced manner. However, noticeable communication overhead was observed, mainly due to IP over the IB protocol. Full utilization of IB Sockets Direct Protocol and the lower latency production version of IBM DAV will reduce this overhead. Copyright © 2011 John Wiley & Sons, Ltd.

Unconventional multi- and many-core processors (e.g., IBM® Cell B.E. ™ and NVIDIA® GPU) have emerged as effective accelerators in trial climate and weather simulations. Yet these climate and weather models typically run on parallel computers with conventional processors (e.g., Intel®, AMD®, and IBM) using Message Passing Interface (MPI). To address challenges involved in efficiently and easily connecting accelerators to parallel computers, we investigated using IBM's Dynamic Application Virtualization (DAV) software in a prototype hybrid computing system with representative climate and weather model components. The hybrid system comprises 2 Intel blades and 2 IBM QS22 Cell B.E. blades, connected with both InfiniBand® (IB) and 1-Gigabit Ethernet. The system significantly accelerates a solar radiation model component by offloading compute-intensive calculations to the Cell blades. Systematic tests show that DAV can seamlessly offload compute-intensive calculations from Intel blades to Cell B.E. blades in a scalable, load-balanced manner. However, noticeable communication overhead was observed, mainly due to IP over IB protocol. Full utilization of IB Sockets Direct Protocol (SDP) and the lower latency production version of DAV will reduce this overhead.