arXiv subjects

Antonios Katsarakis

Publications and source records attributed to Antonios Katsarakis.

9 recordsLinked to original sources

arXiv2026

Flare: Leveraging Serverless Elasticity to Absorb Microservice Load Spikes

Online services strive to maintain application responsiveness even when the traffic is unpredictable and fluctuating. Today's online services are commonly deployed as chains of microservices, each microservice packaged as one or more containers inside virtual machines (VMs). While performant and affordable when the load is steady, VM-based deployments are known to be slow to scale when the load spikes, resulting in degraded performance for end-users of the service. To avoid such performance degradations, service providers can over-provision their deployments; however, such a strategy is costly and inefficient, leaving resources under-utilized for extended periods. To address the challenge of unpredictable load spikes, we propose Flare, a hybrid microservice architecture that combines VMs with serverless computing. Flare utilizes VMs to cost-effectively handle steady workloads and leverages serverless elasticity to absorb traffic spikes. When a spike occurs, Flare detects which specific service(s) are overloaded and shifts the excess load of only those services to serverless, thus minimizing the cost overhead. Flare seamlessly integrates into existing auto-scaling and serverless infrastructure, requiring minimal changes to the control plane and no modifications to the application.

cs.DC↗

arXiv2025

Reliable Replication Protocols on SmartNICs

Today's datacenter applications rely on datastores that are required to provide high availability, consistency, and performance. To achieve high availability, these datastores replicate data across several nodes. Such replication is managed through a reliable protocol designed to keep the replicas consistent using a consistency model, even in the presence of faults. For several applications, strong consistency models are favored over weaker consistency models, as the former guarantee a more intuitive behavior for clients. Furthermore, to meet the demands of high online traffic, datastores must offer high throughput and low latency. However, delivering both strong consistency and high performance simultaneously can be challenging. Reliable replication protocols typically require multiple rounds of communication over the network stack, which introduces latency and increases the load on network resources. Moreover, these protocols consume considerable CPU resources, which impacts the overall performance of applications, especially in high-throughput environments. In this work, we aim to design a hardware-accelerated system for replication protocols to address these challenges. We approach offloading the replication protocol onto SmartNICs, which are specialized network interface cards that can be programmed to implement custom logic directly on the NIC. By doing so, we aim to enhance performance while preserving strong consistency, all while saving valuable CPU cycles that can be used for applications' logic.

cs.DC↗

arXiv2025

TNIC: A Trusted NIC Architecture

We introduce TNIC, a trusted NIC architecture for building trustworthy distributed systems deployed in heterogeneous, untrusted (Byzantine) cloud environments. TNIC builds a minimal, formally verified, silicon root-of-trust at the network interface level. We strive for three primary design goals: (1) a host CPU-agnostic unified security architecture by providing trustworthy network-level isolation; (2) a minimalistic and verifiable TCB based on a silicon root-of-trust by providing two core properties of transferable authentication and non-equivocation; and (3) a hardware-accelerated trustworthy network stack leveraging SmartNICs. Based on the TNIC architecture and associated network stack, we present a generic set of programming APIs and a recipe for building high-performance, trustworthy, distributed systems for Byzantine settings. We formally verify the safety and security properties of our TNIC while demonstrating its use by building four trustworthy distributed systems. Our evaluation of TNIC shows up to 6x performance improvement compared to CPU-centric TEE systems.

cs.CR↗

arXiv2024

vLSM: Low tail latency and I/O amplification in LSM-based KV stores

LSM-based key-value (KV) stores are an important component in modern data infrastructures. However, they suffer from high tail latency, in the order of several seconds, making them less attractive for user-facing applications. In this paper, we introduce the notion of compaction chains and we analyse how they affect tail latency. Then, we show that modern designs reduce tail latency, by trading I/O amplification or require large amounts of memory. Based on our analysis, we present vLSM, a new KV store design that improves tail latency significantly without compromising on memory or I/O amplification. vLSM reduces (a) compaction chain width by using small SSTs and eliminating the tiering compaction required in L0 by modern systems and (b) compaction chain length by using a larger than typical growth factor between L1 and L2 and introducing overlap-aware SSTs in L1. We implement vLSM in RocksDB and evaluate it using db_bench and YCSB. Our evaluation highlights the underlying trade-off among memory requirements, I/O amplification, and tail latency, as well as the advantage of vLSM over current approaches. vLSM improves P99 tail latency by up to 4.8x for writes and by up to 12.5x for reads, reduces cumulative write stalls by up to 60% while also slightly improves I/O amplification at the same memory budget.

cs.DB↗

arXiv2024

DLHT: A Non-blocking Resizable Hashtable with Fast Deletes and Memory-awareness

This paper presents DLHT, a concurrent in-memory hashtable. Despite efforts to optimize hashtables, that go as far as sacrificing core functionality, state-of-the-art designs still incur multiple memory accesses per request and block request processing in three cases. First, most hashtables block while waiting for data to be retrieved from memory. Second, open-addressing designs, which represent the current state-of-the-art, either cannot free index slots on deletes or must block all requests to do so. Third, index resizes block every request until all objects are copied to the new index. Defying folklore wisdom, DLHT forgoes open-addressing and adopts a fully-featured and memory-aware closed-addressing design based on bounded cache-line-chaining. This design offers lock-free index operations and deletes that free slots instantly, (2) completes most requests with a single memory access, (3) utilizes software prefetching to hide memory latencies, and (4) employs a novel non-blocking and parallel resizing. In a commodity server and a memory-resident workload, DLHT surpasses 1.6B requests per second and provides 3.5x (12x) the throughput of the state-of-the-art closed-addressing (open-addressing) resizable hashtable on Gets (Deletes).

cs.DB↗

arXiv2023

Honeycomb: ordered key-value store acceleration on an FPGA-based SmartNIC

In-memory ordered key-value stores are an important building block in modern distributed applications. We present Honeycomb, a hybrid software-hardware system for accelerating read-dominated workloads on ordered key-value stores that provides linearizability for all operations including scans. Honeycomb stores a B-Tree in host memory, and executes SCAN and GET on an FPGA-based SmartNIC, and PUT, UPDATE and DELETE on the CPU. This approach enables large stores and simplifies the FPGA implementation but raises the challenge of data access and synchronization across the slow PCIe bus. We describe how Honeycomb overcomes this challenge with careful data structure design, caching, request parallelism with out-of-order request execution, wait-free read operations, and batching synchronization between the CPU and the FPGA. For read-heavy YCSB workloads, Honeycomb improves the throughput of a state-of-the-art ordered key-value store by at least 1.8x. For scan-heavy workloads inspired by cloud storage, Honeycomb improves throughput by more than 2x. The cost-performance, which is more important for large-scale deployments, is improved by at least 1.5x on these workloads.

cs.DC↗

arXiv2021

Invalidation-Based Protocols for Replicated Datastores

Distributed in-memory datastores underpin cloud applications that run within a datacenter and demand high performance, strong consistency, and availability. A key feature of datastores is data replication. The data are replicated across servers because a single server often cannot handle the request load. Replication is also necessary to guarantee that a server or link failure does not render a portion of the dataset inaccessible. A replication protocol is responsible for ensuring strong consistency between the replicas of a datastore, even when faults occur, by determining the actions necessary to access and manipulate the data. Consequently, a replication protocol also drives the datastore's performance. Existing strongly consistent replication protocols deliver fault tolerance but fall short in terms of performance. Meanwhile, the opposite occurs in the world of multiprocessors, where data are replicated across the private caches of different cores. The multiprocessor regime uses invalidations to afford strongly consistent replication with high performance but neglects fault tolerance. Although handling failures in the datacenter is critical for data availability, we observe that the common operation is fault-free and far exceeds the operation during faults. In other words, the common operating environment inside a datacenter closely resembles that of a multiprocessor. Based on this insight, we draw inspiration from the multiprocessor for high-performance, strongly consistent replication in the datacenter. The primary contribution of this thesis is in adapting invalidating protocols to the nuances of replicated datastores, which include skewed data accesses, fault tolerance, and distributed transactions.

cs.DC↗

arXiv2021

Zeus: Locality-aware Distributed Transactions

State-of-the-art distributed in-memory datastores (FaRM, FaSST, DrTM) provide strongly-consistent distributed transactions with high performance and availability. Transactions in those systems are fully general; they can atomically manipulate any set of objects in the store, regardless of their location. To achieve this, these systems use complex distributed transactional protocols. Meanwhile, many workloads have a high degree of locality. For such workloads, distributed transactions are an overkill as most operations only access objects located on the same server -- if sharded appropriately. In this paper, we show that for these workloads, a single-node transactional protocol combined with dynamic object re-sharding and asynchronously pipelined replication can provide the same level of generality with better performance, simpler protocols, and lower developer effort. We present Zeus, an in-memory distributed datastore that provides general transactions by acquiring all objects involved in the transaction to the same server and executing a single-node transaction on them. Zeus is fault-tolerant and strongly-consistent. At the heart of Zeus is a reliable dynamic object sharding protocol that can move 250K objects per second per server, allowing Zeus to process millions of transactions per second and outperform more traditional distributed transactions on a wide range of workloads that exhibit locality.

cs.DC↗

arXiv2021

Extending Classic Paxos for High-performance Read-Modify-Write Registers

In this work we provide a detailed specification of how we extended and implemented Classic Paxos (CP) to execute Read-Modify-Writes. In addition, we also specify how we implemented All-aboard Paxos over CP and how we use carstamps, to also add ABD reads and writes, to accelerate the common case, where RMWs are not needed. Our specification targets a Key-Value-Store that is deployed within the datacenter, is replicated across 3 to 7 machines and supports reads, writes and RMWs.

cs.DC↗