Search

In our exploration of Composable Memory systems utilizing CXL, we focus on overcoming adoption barriers at Hyperscale, underscored by economic models demonstrating Total Cost of Ownership (TCO). While CXL addresses the pressing memory capacity needs of emerging Hyperscale applications, the escalating demands from evolving use cases such as AI outpace the capabilities of current CXL solutions. Hyperscalers resort to software-based memory (de)compression technology, alleviating memory capacity, storage, and network constraints but incurring a notable "Tax" on Compute CPU cycles. As a pivotal guide to the CXL community, Hyperscalers have formulated the groundbreaking Open Compute Project (OCP) Hyperscale CXL Tiered Memory Expander specification. If implemented, this specification lowers TCO adoption barriers, enabling diverse CXL deployments at both Hyperscaler and Enterprise levels. We present a CXL integrated solution, aligning with the aforementioned specification, introducing an energy-efficient, scalable, hardware-accelerated, Lossless Compressed Memory CXL Tier. This solution, slated for mid-2024 production and open for integration with Memory Expander controller manufacturers, offers 2-3X CXL memory compression in nanoseconds, delivering a 20-25% reduction in TCO for end customers without requiring additional physical slots. In our discussion, we pinpoint areas for collaborative innovation within the CXL Community to expedite software/hardware advancements for CXL Tiered Memory Expansion. Furthermore, we delve into unresolved challenges in Pooled deployment and explore potential solutions, collectively aiming to make CXL adoption a "No Brainer" at Hyperscale., Comment: Presented at the 3rd Workshop on Heterogeneous Composable and Disaggregated Systems (HCDS 2024)

Stream processing applications extract value from raw data through Directed Acyclic Graphs of data analysis tasks. Shared-nothing (SN) parallelism is the de-facto standard to scale stream processing applications. Given an application, SN parallelism instantiates several copies of each analysis task, making each instance responsible for a dedicated portion of the overall analysis, and relies on dedicated queues to exchange data among connected instances. On the one hand, SN parallelism can scale the execution of applications both up and out since threads can run task instances within and across processes/nodes. On the other hand, its lack of sharing can cause unnecessary overheads and hinder the scaling up when threads operate on data that could be jointly accessed in shared memory. This trade-off motivated us in studying a way for stream processing applications to leverage shared memory and boost the scale up (before the scale out) while adhering to the widely-adopted and SN-based APIs for stream processing applications. We introduce STRETCH, a framework that maximizes the scale up and offers instantaneous elastic reconfigurations (without state transfer) for stream processing applications. We propose the concept of Virtual Shared-Nothing (VSN) parallelism and elasticity and provide formal definitions and correctness proofs for the semantics of the analysis tasks supported by STRETCH, showing they extend the ones found in common Stream Processing Engines. We also provide a fully implemented prototype and show that STRETCH's performance exceeds that of state-of-the-art frameworks such as Apache Flink and offers, to the best of our knowledge, unprecedented ultra-fast reconfigurations, taking less than 40 ms even when provisioning tens of new task instances.

The light detection and ranging (LiDAR) technology allows to sense surrounding objects with fine-grained resolution in a large areas. Their data (aka point clouds), generated continuously at very high rates, can provide information to support automated functionality in cyberphysical systems. Clustering of point clouds is a key problem to extract this type of information. Methods for solving the problem in a continuous fashion can facilitate improved processing in e.g. fog architectures, allowing continuous, streaming processing of data close to the sources. We propose Lisco, a single-pass continuous Euclidean-distance-based clustering of LiDAR point clouds, that maximizes the granularity of the data processing pipeline. Besides its algorithmic analysis, we provide a thorough experimental evaluation and highlight its up to 3x improvements and its scalability benefits compared to the baseline, using both real-world datasets as well as synthetic ones to fully explore the worst-cases.

Data streaming relies on continuous queries to process unbounded streams of data in a real-time fashion. It is commonly demanding in computation capacity, given that the relevant applications involve very large volumes of data. Data structures act as articulation points and maintain the state of data streaming operators, potentially supporting high parallelism and balancing the work between them. Prompted by this fact, in this work we study and analyze parallelization needs of these articulation points, focusing on the problem of streaming multiway aggregation, where large data volumes are received from multiple input streams. The analysis of the parallelization needs, as well as of the use and limitations of existing aggregate designs and their data structures, leads us to identify needs for proper shared objects that can achieve low-latency and high throughput multiway aggregation. We present the requirements of such objects as abstract data types and we provide efficient lock-free linearizable algorithmic implementations of them, along with new multiway aggregate algorithmic designs that leverage them, supporting both deterministic order-sensitive and order-insensitive aggregate functions. Furthermore, we point out future directions that open through these contributions. The paper includes an extensive experimental study, based on a variety of aggregation continuous queries on two large datasets extracted from SoundCloud, a music social network, and from a Smart Grid network. In all the experiments, the proposed data structures and the enhanced aggregate operators improved the processing performance significantly, up to one order of magnitude, in terms of both throughput and latency, over the commonly-used techniques based on queues.