Search

We present SupMario, a characterization framework designed to thoroughly analyze, model, and optimize CXL memory performance. SupMario is based on extensive evaluation of 265 workloads spanning 4 real CXL devices within 7 memory latency configurations across 4 processor platforms. SupMario uncovers many key insights, including detailed workload performance at sub-us memory latencies (140-410 ns), CXL tail latencies, CPU tolerance to CXL latencies, CXL performance root-cause analysis and precise performance prediction models. In particular, SupMario performance models rely solely on 12 CPU performance counters and accurately fit over 99% and 91%-94% workloads with a 10% misprediction target for NUMA and CXL memory, respectively. We demonstrate the practical utility of SupMario characterization findings, models, and insights by applying them to popular CXL memory management schemes, such as page interleaving and tiering policies, to identify system inefficiencies during runtime. We introduce a novel ``bestshot'' page interleaving policy and a regulated page tiering policy (Alto) tailored for memory bandwidth- and latency-sensitive workloads. In bandwidth bound scenarios, our ``best-shot'' interleaving, guided by our novel performance prediction model, achieves close-to optimal scenarios by exploiting the aggregate system and CXL/NUMA memory bandwidth. For latency sensitive workloads, Alto, driven by our key insight of utilizing ``amortized'' memory latency to regulate unnecessary page migrations, achieves up to 177% improvement over state-of-the-art memory tiering systems like TPP, as demonstrated through extensive evaluation with 8 real-world applications.

Part segmentation is a crucial task for 3D curvilinear structures like neuron dendrites and blood vessels, enabling the analysis of dendritic spines and aneurysms with scientific and clinical significance. However, their diversely winded morphology poses a generalization challenge to existing deep learning methods, which leads to labor-intensive manual correction. In this work, we propose FreSeg, a framework of part segmentation tasks for 3D curvilinear structures. With Frenet-Frame-based point cloud transformation, it enables the models to learn more generalizable features and have significant performance improvements on tasks involving elongated and curvy geometries. We evaluate FreSeg on 2 datasets: 1) DenSpineEM, an in-house dataset for dendritic spine segmentation, and 2) IntrA, a public 3D dataset for intracranial aneurysm segmentation. Further, we will release the DenSpineEM dataset, which includes roughly 6,000 spines from 69 dendrites from 3 public electron microscopy (EM) datasets, to foster the development of effective dendritic spine instance extraction methods and, consequently, large-scale connectivity analysis to better understand mammalian brains., Comment: 10 pages, 4 figures

We propose quantum methods for solving differential equations that are based on a gradual improvement of the solution via an iterative process, and are targeted at applications in fluid dynamics. First, we implement the Jacobi iteration on a quantum register that utilizes a linear combination of unitaries (LCU) approach to store the trajectory information. Second, we extend quantum methods to Gauss-Seidel iterative methods. Additionally, we propose a quantum-suitable resolvent decomposition based on the Woodbury identity. From a technical perspective, we develop and utilize tools for the block encoding of specific matrices as well as their multiplication. We benchmark the approach on paradigmatic fluid dynamics problems. Our results stress that instead of inverting large matrices, one can program quantum computers to perform multigrid-type computations and leverage corresponding advances in scientific computing., Comment: 8 pages, 7 figures

The Compute Express Link (CXL) is an open industry-standard interconnect between processors and devices such as accelerators, memory buffers, smart network interfaces, persistent memory, and solid-state drives. CXL offers coherency and memory semantics with bandwidth that scales with PCIe bandwidth while achieving significantly lower latency than PCIe. All major CPU vendors, device vendors, and datacenter operators have adopted CXL as a common standard. This enables an inter-operable ecosystem that supports key computing use cases including highly efficient accelerators, server memory bandwidth and capacity expansion, multi-server resource pooling and sharing, and efficient peer-to-peer communication. This survey provides an introduction to CXL covering the standards CXL 1.0, CXL 2.0, and CXL 3.0. We further survey CXL implementations, discuss CXL's impact on the datacenter landscape, and future directions.

Some faults in data center networks require hours to days to repair because they may need reboots, re-imaging, or manual work by technicians. To reduce traffic impact, cloud providers \textit{mitigate} the effect of faults, for example, by steering traffic to alternate paths. The state-of-art in automatic network mitigations uses simple safety checks and proxy metrics to determine mitigations. SWARM, the approach described in this paper, can pick orders of magnitude better mitigations by estimating end-to-end connection-level performance (CLP) metrics. At its core is a scalable CLP estimator that quickly ranks mitigations with high fidelity and, on failures observed at a large cloud provider, outperforms the state-of-the-art by over 700$\times$ in some cases.

Recent work shows the effectiveness of Machine Learning (ML) to reduce cache miss ratios by making better eviction decisions than heuristics. However, state-of-the-art ML caches require many predictions to make an eviction decision, making them impractical for high-throughput caching systems. This paper introduces Machine learning At the Tail (MAT), a framework to build efficient ML-based caching systems by integrating an ML module with a traditional cache system based on a heuristic algorithm. MAT treats the heuristic algorithm as a filter to receive high-quality samples to train an ML model and likely candidate objects for evictions. We evaluate MAT on 8 production workloads, spanning storage, in-memory caching, and CDNs. The simulation experiments show MAT reduces the number of costly ML predictions-per-eviction from 63 to 2, while achieving comparable miss ratios to the state-of-the-art ML cache system. We compare a MAT prototype system with an LRU-based caching system in the same setting and show that they achieve similar request rates.

Public cloud providers seek to meet stringent performance requirements and low hardware cost. A key driver of performance and cost is main memory. Memory pooling promises to improve DRAM utilization and thereby reduce costs. However, pooling is challenging under cloud performance requirements. This paper proposes Pond, the first memory pooling system that both meets cloud performance goals and significantly reduces DRAM cost. Pond builds on the Compute Express Link (CXL) standard for load/store access to pool memory and two key insights. First, our analysis of cloud production traces shows that pooling across 8-16 sockets is enough to achieve most of the benefits. This enables a small-pool design with low access latency. Second, it is possible to create machine learning models that can accurately predict how much local and pool memory to allocate to a virtual machine (VM) to resemble same-NUMA-node memory performance. Our evaluation with 158 workloads shows that Pond reduces DRAM costs by 7% with performance within 1-5% of same-NUMA-node VM allocations., Comment: Update affiliations

Cloud platforms run many software agents on each server node. These agents manage all aspects of node operation, and in some cases frequently collect data and make decisions. Unfortunately, their behavior is typically based on pre-defined static heuristics or offline analysis; they do not leverage on-node machine learning (ML). In this paper, we first characterize the spectrum of node agents in Azure, and identify the classes of agents that are most likely to benefit from on-node ML. We then propose SOL, an extensible framework for designing ML-based agents that are safe and robust to the range of failure conditions that occur in production. SOL provides a simple API to agent developers and manages the scheduling and running of the agent-specific functions they write. We illustrate the use of SOL by implementing three ML-based agents that manage CPU cores, node power, and memory placement. Our experiments show that (1) ML substantially improves our agents, and (2) SOL ensures that agents operate safely under a variety of failure conditions. We conclude that ML-based agents show significant potential and that SOL can help build them.

Redy is a cloud service that provides high performance caches using RDMA-accessible remote memory. An application can customize the performance of each cache with a service level objective (SLO) for latency and throughput. By using remote memory, it can leverage stranded memory and spot VM instances to reduce the cost of its caches and improve data center resource utilization. Redy automatically customizes the resource configuration for the given SLO, handles the dynamics of remote memory regions, and recovers from failures. The experimental evaluation shows that Redy can deliver its promised performance and robustness under remote memory dynamics in the cloud. We augment a production key-value store, FASTER, with a Redy cache. When the working set exceeds local memory, using Redy is significantly faster than spilling to SSDs., Comment: This is the extended report of Redy (accepted at VLDB 2022)

Continuously monitoring a wide variety of performance and fault metrics has become a crucial part of operating large-scale datacenter networks. In this work, we ask whether we can reduce the costs to monitor -- in terms of collection, storage and analysis -- by judiciously controlling how much and which measurements we collect. By positing that we can treat almost all measured signals as sampled time-series, we show that we can use signal processing techniques such as the Nyquist-Shannon theorem to avoid wasteful data collection. We show that large savings appear possible by analyzing tens of popular measurements from a production datacenter network. We also discuss the technical challenges that must be solved when applying these techniques in practice.

This article contains data on structural characterization of the [C2Mim][NTf2] in bulk and in nano-confined environment obtained using MD simulations. These data supplement those presented in the paper Insights from Molecular Dynamics Simulations on Structural Organization and Diffusive Dynamics of an Ionic Liquid at Solid and Vacuum Interfaces, where force fields with three different charge methods and three charge scaling factors were used for the analysis of the IL in the bulk, at the interface with the vacuum and the IL film in the contact with a hydroxylated alumina surface. Here, we present details on the construction of the model systems in an extended detailed methods section. Furthermore, for best parametrization, structural and dynamic properties of IL in different environment are studied with certain features presented herein., Comment: 13 pages, 12 figures. arXiv admin note: substantial text overlap with arXiv:1903.09450

Hypothesis A prototypical modelling approach is required for a full characterisation of the static and equilibrium dynamical properties of confined ionic liquids (ILs), in order to gain predictive power of properties that are difficult to extract from experiments. Such a protocol needs to be constructed by benchmarking molecular dynamics simulations against available experiments. Simulations We perform an in-depth study of [C2Mim][NTf2] in bulk, at the vacuum and at hydroxylated alumina surface. Using the charge methods CHelpG, RESP-HF and RESP-B3LYP with charge scaling factors 1.0, 0.9 and 0.85, we search for an optimum non-polarizable force field by benchmarking against self-diffusion coefficients, surface tension, X-ray reflectivity data, and structural data. Findings Benchmarking, which relies on establishing the significance of an appropriate size of the model systems and the length of the simulations, yields RESP-HF/0.9 as the best suited force field for this IL overall. A complete and accurate characterisation of the spatially-dependent internal configurational space and orientation of IL molecules relative to the solid and vacuum interfaces is obtained. Furthermore, the density and mobility of IL ions in the plane parallel and normal to the interfaces is evaluated and the correlation between the stratification and dynamics in the interfacial layers is detectable deep into the films., Comment: 14 pages, 9 figures in main text and 14 figures in Supporting Information

Synaptic connectivity detection is a critical task for neural reconstruction from Electron Microscopy (EM) data. Most of the existing algorithms for synapse detection do not identify the cleft location and direction of connectivity simultaneously. The few methods that computes direction along with contact location have only been demonstrated to work on either dyadic (most common in vertebrate brain) or polyadic (found in fruit fly brain) synapses, but not on both types. In this paper, we present an algorithm to automatically predict the location as well as the direction of both dyadic and polyadic synapses. The proposed algorithm first generates candidate synaptic connections from voxelwise predictions of signed proximity generated by a 3D U-net. A second 3D CNN then prunes the set of candidates to produce the final detection of cleft and connectivity orientation. Experimental results demonstrate that the proposed method outperforms the existing methods for determining synapses in both rodent and fruit fly brain.

Many catalyst devices employ flow through porous structures, which leads to a complex macroscopic mass and heat transport. To unravel the detailed dynamics of the reactive gas flow, we present an all-encompassing model, consisting of thermal lattice Boltzmann model by Kang et al., used to solve the heat and mass transport in the gas domain, coupled to a finite differences solver for the heat equation in the solid via thermal reactive boundary conditions for a consistent treatment of the reaction enthalpy. The chemical surface reactions are incorporated in a flexible fashion through flux boundary conditions at the gas-solid interface. We scrutinize the thermal FD-LBM by benchmarking the macroscopic transport in the gas domain as well as conservation of the enthalpy across the solid-gas interface. We exemplify the applicability of our model by simulating the reactive gas flow through a microporous material catalysing the so-called water-gas-shift reaction.

Many recent caching systems aim to improve miss ratios, but there is no good sense among practitioners of how much further miss ratios can be improved. In other words, should the systems community continue working on this problem? Currently, there is no principled answer to this question. In practice, object sizes often vary by several orders of magnitude, where computing the optimal miss ratio (OPT) is known to be NP-hard. The few known results on caching with variable object sizes provide very weak bounds and are impractical to compute on traces of realistic length. We propose a new method to compute upper and lower bounds on OPT. Our key insight is to represent caching as a min-cost flow problem, hence we call our method the flow-based offline optimal (FOO). We prove that, under simple independence assumptions, FOO's bounds become tight as the number of objects goes to infinity. Indeed, FOO's error over 10M requests of production CDN and storage traces is negligible: at most 0.3%. FOO thus reveals, for the first time, the limits of caching with variable object sizes. While FOO is very accurate, it is computationally impractical on traces with hundreds of millions of requests. We therefore extend FOO to obtain more efficient bounds on OPT, which we call practical flow-based offline optimal (PFOO). We evaluate PFOO on several full production traces and use it to compare OPT to prior online policies. This analysis shows that current caching systems are in fact still far from optimal, suffering 11-43% more cache misses than OPT, whereas the best prior offline bounds suggest that there is essentially no room for improvement.

Realistic oxide materials are often semiconductors, in particular at elevated temperatures, and their surfaces contain undercoordiated atoms at structural defects such as steps and corners. Using hybrid density-functional theory and ab initio atomistic thermodynamics, we investigate the interplay of bond-making, bond-breaking, and charge-carrier trapping at the corner defects at the (100) surface of a p-doped MgO in thermodynamic equilibrium with an O2 atmosphere. We show that by manipulating the coordination of surface atoms one can drastically change and even reverse the order of stability of reduced versus oxidized surface sites., Comment: 5 papges, 4 figures

Reconstructing a synaptic wiring diagram, or connectome, from electron microscopy (EM) images of brain tissue currently requires many hours of manual annotation or proofreading (Kasthuri and Lichtman, 2010; Lichtman and Sanes, 2008; Seung, 2009). The desire to reconstruct ever larger and more complex networks has pushed the collection of ever larger EM datasets. A cubic millimeter of raw imaging data would take up 1 PB of storage and present an annotation project that would be impractical without relying heavily on automatic segmentation methods. The RhoanaNet image processing pipeline was developed to automatically segment large volumes of EM data and ease the burden of manual proofreading and annotation. Based on (Kaynig et al., 2015), we updated every stage of the software pipeline to provide better throughput performance and higher quality segmentation results. We used state of the art deep learning techniques to generate improved membrane probability maps, and Gala (Nunez-Iglesias et al., 2014) was used to agglomerate 2D segments into 3D objects. We applied the RhoanaNet pipeline to four densely annotated EM datasets, two from mouse cortex, one from cerebellum and one from mouse lateral geniculate nucleus (LGN). All training and test data is made available for benchmark comparisons. The best segmentation results obtained gave $V^\text{Info}_\text{F-score}$ scores of 0.9054 and 09182 for the cortex datasets, 0.9438 for LGN, and 0.9150 for Cerebellum. The RhoanaNet pipeline is open source software. All source code, training data, test data, and annotations for all four benchmark datasets are available at www.rhoana.org., Comment: 13 pages, 4 figures

Graphene and carbon nanotubes have extraordinary mechanical and electronic properties. Intrinsic line defects such as local non-hexagonal reconstructions or grain boundaries, however, significantly reduce the tensile strength, but feature exciting electronic properties. Here, we address the properties of line defects in graphene from first-principles on the level of full-potential density-functional theory, and assess doping as one strategy to strengthen such materials. We carefully disentangle the global and local effect of doping by comparing results from the virtual crystal approximation with those from local substitution of chemical species, in order to gain a detailed understanding of the breaking and stabilization mechanisms. We find that n-type doping or local substitution with nitrogen increases the ultimate tensile strength significantly. In particular, it can stabilize the defects beyond the ultimate tensile strength of the pristine material. We therefore propose this as a key strategy to strengthen graphenic materials. Furthermore, we find that doping and/or applying external stress lead to tunable and technologically interesting metal/semi-conductor transitions.

We perform full-potential screened-hybrid density-functional theory (DFT) calculations to compare the thermodynamic stability of neutral and charged states of the surface oxygen vacancy at the rutile TiO$_2$(110) surface. Solid-state (QM/MM) embedded-cluster calculations are employed to account for the strong TiO$_2$ polarization response to the charged defect states. Similar to the situation for the bulk O vacancy, the +2 charge state $V_{\rm O}^{2+}$ is found to be energetically by far most stable. Only for Fermi-level positions very close to the conduction band, small polarons may at best be trapped by the charged vacancy. The large decrease of the $V_{\rm O}^{2+}$ formation energy with decreasing Fermi-level position indicates strongly enhanced surface O vacancy concentrations for $p$-doped samples.

Let $\mathrm{\mathbf{L}\rightarrow \mathbf{X}}$ be a base point free $\mathrm{\mathbb{T}=T^{\mathbb{C}}}$-linearized hermitian line bundle over a compact variety $\mathrm{\mathbf{X}}$ where $\mathrm{T=\left(S^{1}\right)^{m}}$ is a real torus. The main focus of this paper is to describe the asymptotic behavior of a certain class of sequences $\mathrm{\left(s_{n}\right)_{n}}$ of $\mathrm{\mathbb{T}}$-eigensections $\mathrm{s_{n}\in H^{0}\left(\mathbf{X},\mathbf{L}^{n}\right)}$ as $\mathrm{n\rightarrow \infty}$, introduced by Shiffman, Tate and Zelditch, and its connection to the geometry of the Hilbert quotient $\mathrm{\pi\!:\!\bf{X}^{ss}_{\xi}\rightarrow \mathbf{X}^{ss}_{\xi}/\!\!/\mathbb{T}}$ where $\mathrm{\xi\in \mathfrak{t}^{*}}$., Comment: 67 pages - This work is the author's doctoral thesis submitted in partial fulfillment of the requirements for the degree of Doktor rer. nat. at the Ruhr-Universit\"at Bochum