arXiv subjects

Tobias Ziegler

Publications and source records attributed to Tobias Ziegler.

13 recordsLinked to original sources

arXiv2026

High-Performance DBMSs with io_uring: When and How to use it

We study how modern database systems can leverage the Linux io_uring interface for efficient, low-overhead I/O. io_uring is an asynchronous system call batching interface that unifies storage and network operations, addressing limitations of existing Linux I/O interfaces. However, naively replacing traditional I/O interfaces with io_uring does not necessarily yield performance benefits. To demonstrate when io_uring delivers the greatest benefits and how to use it effectively in modern database systems, we evaluate it in two use cases: Integrating io_uring into a storage-bound buffer manager and using it for high-throughput data shuffling in network-bound analytical workloads. We further analyze how advanced io_uring features, such as registered buffers and passthrough I/O, affect end-to-end performance. Our study shows when low-level optimizations translate into tangible system-wide gains and how architectural choices influence these benefits. Building on these insights, we derive practical guidelines for designing I/O-intensive systems using io_uring and validate their effectiveness in a case study of PostgreSQL's recent io_uring integration, where applying our guidelines yields a performance improvement of 14%.

cs.DB↗

arXiv2026

BtrLog: Low-Latency Logging for Cloud Database Systems

Cloud database systems cannot rely on instance-local disks for write-ahead logging (WAL) durability, forcing WAL onto remote storage. Existing options are unsatisfying: remote block storage like EBS is easy to adopt but adds substantial write latency and cost, while object storage offers excellent durability and low storage cost but is impractical for OLTP due to high latency and per-append cost. Many cloud-native databases, therefore, depend on purpose-built logging backends, which are typically proprietary and tightly coupled to engine-specific replication and recovery protocols, limiting reuse. We present BtrLog, a reusable cloud logging service that combines low-latency durable appends with low-cost archival for the common single-writer architecture. BtrLog replicates log records across a quorum of SSD-backed log nodes in a single network round trip, reducing sensitivity to stragglers in commit latency. To minimize storage cost, log nodes archive records to object storage as large segments, which are written asynchronously and off the latency-critical write path. In our evaluation, BtrLog achieves lower latency than EBS and enables higher end-to-end transaction throughput when integrated into a DBMS.

cs.DB↗

arXiv2026

The Bi-Channel Networking Paradigm for Database Systems in the Cloud

When network links were slow, cloud and distributed database systems could rely on generic kernel abstractions and treat network communication as a black box. With today's fast cloud networks, this approach breaks down: database performance becomes limited by the CPU overhead of the kernel TCP stack. Replacing TCP with user-space UDP can reduce this overhead, but it requires reimplementing essential guarantees, such as reliability and ordering. To solve this conundrum, database systems should no longer treat networking as a black box but co-design it with database operations. We propose the bi-channel paradigm for database systems, which separates communication into two channels: A high-performance data path for latency- and bandwidth-sensitive operations, and a reliable control path for coordination and recovery. We implement the paradigm by combining user-space UDP and kernel-based TCP, though other stack combinations are possible. This design exploits modern NIC capabilities while preserving TCP's reliability. We demonstrate the paradigm's efficiency and simplicity in two representative settings: a distributed shuffle saturating 200 Gbit/s with three CPU cores, and a replicated key-value store processing millions of messages per second.

cs.DB↗

arXiv2026

How to Write to SSDs

This paper demonstrates that adopting out-of-place writes is essential for database systems to fully leverage SSD performance and extend SSD lifespan. We propose a set of out-of-place optimizations that collectively reduce write amplification across both the DBMS and SSD layers. We redesign the in-place, B-tree-based LeanStore to write out-of-place and support these optimizations, and evaluate it on diverse OLTP benchmarks, dataset sizes, and SSDs. The final design improves throughput by 1.65-2.24x and reduces flash writes per operation by 6.2-9.8x on YCSB-A. On TPC-C with 15,000 warehouses, throughput improves by 2.45x while flash writes decrease by 7.2x. Finally, we show that the architecture can seamlessly support novel SSD interfaces such as ZNS and FDP.

cs.DB↗

arXiv2025

X-TIME: An in-memory engine for accelerating machine learning on tabular data with CAMs

Structured, or tabular, data is the most common format in data science. While deep learning models have proven formidable in learning from unstructured data such as images or speech, they are less accurate than simpler approaches when learning from tabular data. In contrast, modern tree-based Machine Learning (ML) models shine in extracting relevant information from structured data. An essential requirement in data science is to reduce model inference latency in cases where, for example, models are used in a closed loop with simulation to accelerate scientific discovery. However, the hardware acceleration community has mostly focused on deep neural networks and largely ignored other forms of machine learning. Previous work has described the use of an analog content addressable memory (CAM) component for efficiently mapping random forests. In this work, we develop an analog-digital architecture that implements a novel increased precision analog CAM and a programmable chip for inference of state-of-the-art tree-based ML models, such as XGBoost, CatBoost, and others. Thanks to hardware-aware training, X-TIME reaches state-of-the-art accuracy and 119x higher throughput at 9740x lower latency with >150x improved energy efficiency compared with a state-of-the-art GPU for models with up to 4096 trees and depth of 8, with a 19W peak power consumption.

cs.LG↗

arXiv2024

Synaptogen: A cross-domain generative device model for large-scale neuromorphic circuit design

We present a fast generative modeling approach for resistive memories that reproduces the complex statistical properties of real-world devices. To enable efficient modeling of analog circuits, the model is implemented in Verilog-A. By training on extensive measurement data of integrated 1T1R arrays (6,000 cycles of 512 devices), an autoregressive stochastic process accurately accounts for the cross-correlations between the switching parameters, while non-linear transformations ensure agreement with both cycle-to-cycle (C2C) and device-to-device (D2D) variability. Benchmarks show that this statistically comprehensive model achieves read/write throughputs exceeding those of even highly simplified and deterministic compact models.

cs.NE↗

arXiv2023

Work-in-Progress: A Universal Instrumentation Platform for Non-Volatile Memories

Emerging non-volatile memories (NVMs) represent a disruptive technology that allows a paradigm shift from the conventional von Neumann architecture towards more efficient computing-in-memory (CIM) architectures. Several instrumentation platforms have been proposed to interface NVMs allowing the characterization of single cells and crossbar structures. However, these platforms suffer from low flexibility and are not capable of performing CIM operations on NVMs. Therefore, we recently designed and built the NeuroBreakoutBoard, a highly versatile instrumentation platform capable of executing CIM on NVMs. We present our preliminary results demonstrating a relative error < 5% in the range of 1 k$Ω$ to 1 M$Ω$ and showcase the switching behavior of a HfO$_2$/Ti-based memristive cell.

cs.AR↗

arXiv2023

Optical link acquisition for the LISA mission with in-field pointing architecture

We present a comprehensive simulation of the spatial acquisition of optical links for the LISA mission in the in-field pointing architecture, where a fast pointing mirror is used to move the field-of-view of the optical transceiver, which was studied as an alternative scheme to the baselined telescope pointing architecture. The simulation includes a representative model of the far-field intensity distribution and the beam detection process using a realistic detector model, and a model of the expected platform jitter for two alternative control modes with different associated jitter spectra. For optimally adjusted detector settings and accounting for the actual far-field beam profile, we investigate the dependency of acquisition performance on the jitter spectrum and the track-width of the search spiral, while scan speed and detector integration time are varied over several orders of magnitude. Results show a strong dependency of the probability for acquisition failure on the width of the auto-correlation function of the jitter spectrum, which we compare to predictions of analytical models. Depending on the choice of scan speed, three different regimes may be entered which differ in failure probability by several orders of magnitude. We then use these results to optimize the acquisition architecture for the given jitter spectra with respect to failure rate and overall duration, concluding that the full constellation could be acquired on average in less than one minute. Our method and findings can be applied to any other space mission using a fine-steering mirror for link acquisition.

physics.ins-det↗

arXiv2022

LISA Point-Ahead Angle Control for Optimal Tilt-to-Length Noise Estimation

The Laser Interferometer Space Antenna (LISA) mission features a three-spacecraft long-arm constellation intended to detect gravitational wave sources in the low-frequency band up to 1 Hz via laser interferometry. The paper presents an open-loop control strategy for point-ahead angle (PAA) correction required to maintain the optical links of the moving constellation. The control strategy maximizes periods between adjustments at the constellation level and is shown to be optimal from the perspective of estimating and correcting tilt-to-length (TTL) coupling. TTL is a noise source that couples angular spacecraft jitter and jitter of optical subassemblies with longitudinal interferometer measurements. Without precise TTL noise estimation and correction, TTL coupling fundamentally limits the detector's sensitivity.

physics.ins-det↗

arXiv2022

P4DB -- The Case for In-Network OLTP (Extended Technical Report)

In this paper we present a new approach for distributed DBMSs called P4DB, that uses a programmable switch to accelerate OLTP workloads. The main idea of P4DB is that it implements a transaction processing engine on top of a P4-programmable switch. The switch can thus act as an accelerator in the network, especially when it is used to store and process hot (contended) tuples on the switch. In our experiments, we show that P4DB hence provides significant benefits compared to traditional DBMS architectures and can achieve a speedup of up to 8x.

cs.DB↗

arXiv2014

The vectorial photoelectric effect under solar irradiance and its application to sun sensing

Sun sensors are an integral part of the attitude and orbit control system onboard almost any spacecraft. While the majority of standard analogue sun sensors is based on photo-detectors which produce photo-currents proportional to the cosine of the incidence angle (cosine detectors), we propose an alternative scheme where the vectorial photoelectric effect is exploited to achieve a higher sensitivity of the sensed photo-current to the incidence angle. The vectorial photo-effect is investigated in detail for metal cathode detectors in a space environment. Besides long operational lifetimes without significant degradation, metal cathode detectors are insensitive to earth albedo, which may significantly reduce the errors affecting attitude measurements in low earth orbits. Sensitivity curves are calculated and trade-offs performed with the aim of optimizing the sensitivity whilst also providing currents sufficient for detection. Simple applications and detector configurations are also discussed and compared to existing designs.

physics.ins-det↗

arXiv2012

Modeling and Performance of Contact-Free Discharge Systems for Space Inertial Sensors

This article presents a detailed overview and assessment of contact-free UV light discharge systems (UVDS) needed to control the variable electric charge level of free-flying test masses which are part of high precision inertial sensors in space. A comprehensive numerical analysis approach on the basis of experimental data is detailed. This includes UV light ray tracing, the computation of time variant electric fields inside the complex inertial sensor geometry, and the simulation of individual photo-electron trajectories. Subsequent data analysis allows to determine key parameters to set up an analytical discharge model. Such a model is an essential system engineering tool needed for requirement breakdown and subsystem specification, performance budgeting, on-board charge control software development, and instrument modeling within spacecraft end-to-end performance simulators. Different types of UVDS design concepts are presented and assessed regarding their robustness and performance. Critical hardware aspects like electron emission from air-contaminated surfaces, interfaces with other subsystems, and spacecraft operations are considered. The focus is on the modeling and performance evaluation of the existing UVDS on board LISA Pathfinder, an ESA technology demonstrator spacecraft to be launched in 2014. The results have motivated the design of a more robust discharge system concept for cubical test mass inertial sensors for future space missions. The developed analysis tools have been used for design optimization and performance assessment of the proposed design. A significant improvement of relevant robustness and performance figures has been achieved.

physics.ins-det↗

arXiv2012

Energy distribution and quantum yield for photoemission from air-contaminated gold surfaces under UV illumination close to the threshold

The kinetic energy distributions of photo-electrons emitted from gold surfaces under illumination by UV-light close to the threshold are measured and analyzed. Samples are prepared as chemically clean through Ar-Ion sputtering and then exposed to atmosphere for variable durations before Quantum Yield measurements are performed after evacuation. During measurements the bias voltage applied to the sample is varied and the resulting emission current measured. Taking the derivative of the current-voltage curve yields the energy distribution which is found to closely resemble the distribution of total energies derived by DuBridge for emission from a free electron gas. We investigate the dependence of distribution shape and width on electrode geometry and contaminant substances adsorbed from the atmosphere, in particular to water and hydro-carbons. Emission efficiency increases initially during air exposure before diminishing to zero on a timescale of several hours, whilst subsequent annealing of the sample restores emissivity. A model fit function, in good quantitative agreement with the measured data, is introduced which accounts for the experiment-specific electrode geometry and an energy dependent transmission coefficient. The impact of large patch potential fields from contact potential drops between sample and sample holder is investigated. The total quantum yield is split into bulk and surface contributions which are tested for their sensitivity to light incidence angle and polarization. Our results are directly applicable to model parameters for the contact-free discharge system onboard the LISA Pathfinder spacecraft.

cond-mat.mtrl-sci↗