arXiv subjects

Sandeep Kulkarni

Publications and source records attributed to Sandeep Kulkarni.

At least 19 recordsLinked to original sources

arXiv2026

Domain Extension of Lock-Freedom and Wait-Freedom for Group Computations

A domain extension of a definition refers to broadening the scope of a definition so that it applies to a larger set of cases than originally specified. The notion of lock-free and wait-free computation is designed for the domain of tasks that are completed by a single thread (in competition with other threads). The goal of this paper is to extend the definition of lock-freedom and wait-freedom to group-computations (denoted by gl-freedom and gw-freedom) that require that the task at hand must be completed by a collaboration between multiple threads. When extending a definition, certain constraints must be respected: the new domain must remain logically consistent with the original meaning, the extension should not introduce contradictions or ambiguities, and it must preserve the essential properties that make the definition valid and useful. We demonstrate this by showing that our extended definition is consistent with the original definition when the group consists of a single thread. We note that extension allows us to characterize programs in a new domain (distributed computing, NUMA computation systems, systems with private data for different threads, etc.) instead of relegating them to be in the same category (deadlock/livelock-free) without regard to the actual properties of that program. We also illustrate this definition with various examples.

cs.DC↗

arXiv2025

Lock-free Asynchronously Distributed Linked Lists

Modern databases use dynamic search structures that store an enormous amount of data, and often serve them using multi-threaded algorithms to support the ever-increasing throughput needs. When this throughput need exceeds the capacity of the machine hosting the structure, one either needs to replace the underlying hardware (an option that is typically not viable and introduces a long down time) or make the data structure distributed. Static partitioning of the data structure for distribution is not desirable, as it is prone to uneven load distribution over time, and having to change the partitioning scheme later will require downtime. The goal of this paper is to extend a concurrent data structure to distributed data structures that provide dynamic load balancing while preserving important properties such as lock freedom. With this intuition, first, we introduce the notion of conditional lock-freedom which extends the notion of lock-free computation with reasonable assumptions about communication between processes. Then, we present DiLi, a conditional lock-free, linearizable, and distributed linked list that can be asynchronously and dynamically (1) partitioned into multiple sublists and (2) load balanced by distributing sublists across multiple machines. DiLi contains primitives for these that also maintain the lock-free property of the underlying search structure that supports find, remove, and insert of a key as the client operations. We show that DiLi bridges the gap between concurrent data structures and distributed data structures. Specifically, DiLi provides comparable (and better in write-intensive workloads) performance to skip lists (which are typically the fastest data structures for search in a concurrent environment). In addition, it provides horizontal scaling with dynamic load balancing in a distributed environment.

cs.DC↗

arXiv2025

Asynchronous Checkpoint for Eventually Consistent Databases

We focus on the problem of checkpointing (or taking a snapshot) in fully replicated eventually consistent distributed databases. In particular, we consider the problem of taking Distributed Transaction-Consistent Snapshots (DTCS). A typical example of such a system is a replicated main-memory database that provides strong eventual consistency. This problem is important and challenging for several reasons: (1) eventual consistency often creates anomalies that the users do not anticipate. Hence, frequent snapshots that can be used to ascertain desired invariants are highly beneficial in their maintenance, and (2) traditional distributed snapshot algorithms lead to significant overhead and/or inconsistencies such as storing dirty writes of incomplete transactions. A key benefit of DTCS is that it summarizes the computation by a sequence of snapshots that are strongly consistent even though the underlying computation is only weakly consistent. In essence, when anomalies arise in an eventually consistent system, DTCS enables one to concentrate solely on the snapshots surrounding the time point of the anomaly. By showing that traditional distributed snapshots lead to inconsistencies and/or excessive overhead, we define the notion of size-minimal DTCS for fully replicated databases. We present MuFASA, an algorithm for a size-minimal DTCS with minimal checkpointing overhead (only O(n) new messages and the addition of a single counter for existing messages). MuFASA also provides a significant benefit over existing checkpointing algorithms for distributed systems and replicated main-memory databases by being a fully asynchronous protocol.

cs.DC↗

arXiv2025

Fault-Tolerant Decentralized Distributed Asynchronous Federated Learning with Adaptive Termination Detection

Federated Learning (FL) facilitates collaborative model training across distributed clients while ensuring data privacy. Traditionally, FL relies on a centralized server to coordinate learning, which creates bottlenecks and a single point of failure. Decentralized FL architectures eliminate the need for a central server and can operate in either synchronous or asynchronous modes. Synchronous FL requires all clients to compute updates and wait for one another before aggregation, guaranteeing consistency but often suffering from delays due to slower participants. Asynchronous FL addresses this by allowing clients to update independently, offering better scalability and responsiveness in heterogeneous environments. Our research develops an asynchronous decentralized FL approach in two progressive phases. (a) In Phase 1, we develop an asynchronous FL framework that enables clients to learn and update independently, removing the need for strict synchronization. (b) In Phase 2, we extend this framework with fault tolerance mechanisms to handle client failures and message drops, ensuring robust performance even under unpredictable conditions. As a central contribution, we propose Client-Confident Convergence and Client-Responsive Termination novel techniques that provide each client with the ability to autonomously determine appropriate termination points. These methods ensure that all active clients conclude meaningfully and efficiently, maintaining reliable convergence despite the challenges of asynchronous communication and faults.

cs.DC↗

arXiv2024

Distributing Context-Aware Shared Memory Data Structures: A Case Study on Singly-Linked Lists

In this paper, we study the partitioning of a context-aware shared memory data structure so that it can be implemented as a distributed data structure running on multiple machines. By context-aware data structures, we mean that the result of an operation not only depends upon the value of the shared data but also upon the previous operations performed by the same client. While there is substantial work on designing distributed data structures, designing distributed context-aware data structures has not received much attention. We focus on singly-linked lists as a case study of the context-aware data structure. We start with a shared memory context-aware lock-free singly-linked list and show how it can be transformed into a distributed lock-free context-aware singly-linked list. The main challenge in such a transformation is to preserve properties of client-visible operations of the underlying data structure. We present two protocols that preserve these properties of client-visible operations of the linked list. In the first protocol, the distribution is done in the background as a low priority task, while in the second protocol the client-visible operations help the task of distribution without affecting client latency. In both protocols, the client-visible operations remain lock-free. Also, our transformation approach does not utilize any hardware primitives (except a compare-and-swap operation on a single word). We note that our transformation is generic and can be used for other lock-free context-aware data structures that can be constructed from singly-linked lists.

cs.DC↗

arXiv2022

Impact of the Consistency Model on Checkpointing of Distributed Shared Memory

In this report, we consider the impact of the consistency model on checkpointing and rollback algorithms for distributed shared memory. In particular, we consider specific implementations of four consistency models for distributed shared memory, namely, linearizability, sequential consistency, causal consistency and eventual consistency, and develop checkpointing and rollback algorithms that can be integrated into the implementations of the consistency models. Our results empirically demonstrate that the mechanisms used to implement stronger consistency models lead to simpler or more efficient checkpointing algorithms.

cs.DC↗

arXiv2020

Efficient Two-Layered Monitor for Partially Synchronous Distributed Systems (Technical Report)

Monitoring distributed systems to ensure their correctness is a challenging and expensive but essential problem. It is challenging because while execution of a distributed system creates a partial order among events, the monitor will typically observe only one serialization of that partial order. This means that even if the observed serialization is consistent with the system specifications, the monitor cannot assume that the system is correct because some other unobserved serialization can be inconsistent with the system specifications. Existing solutions that guarantee identification of all such unobserved violations require some combination of lots of time and large clocks, e.g. O(n) sized Vector Clocks. We present a new, efficient two-layered monitoring approach that overcomes both the time and space limitations of earlier monitors. The first layer is imprecise but efficient and the second layer is precise but (relatively) inefficient. We show that the combination of these two layers reduces the cost of monitoring by 85-95%. Furthermore, the two-layered monitor permits the use of O(1) sized Hybrid Logical Clocks.

cs.DC↗

arXiv2019

Achieving Starvation-Freedom with Greater Concurrency in Multi-Version Object-based Transactional Memory Systems

To utilize the multi-core processors properly concurrent programming is needed. Concurrency control is the main challenge while designing a correct and efficient concurrent program. Software Transactional Memory Systems (STMs) provides ease of multithreading to the programmer without worrying about concurrency issues such as deadlock, livelock, priority inversion, etc. Most of the STMs works on read-write operations known as RWSTMs. Some STMs work at high-level operations and ensure greater concurrency than RWSTMs. Such STMs are known as Object-Based STMs (OSTMs). The transactions of OSTMs can return commit or abort. Aborted OSTMs transactions retry. But in the current setting of OSTMs, transactions may starve. So, we proposed a Starvation-Free OSTM (SF-OSTM) which ensures starvation-freedom in object based STM systems while satisfying the correctness criteria as co-opacity. Databases, RWSTMs and OSTMs say that maintaining multiple versions corresponding to each key of transaction reduces the number of aborts and improves the throughput. So, to achieve greater concurrency, we proposed Starvation-Free Multi-Version OSTM (SF-MVOSTM) which ensures starvation-freedom while storing multiple versions corresponding to each key and satisfies the correctness criteria such as local opacity. To show the performance benefits, We implemented three variants of SF-MVOSTM (SF-MVOSTM, SF-MVOSTM-GC and SF-KOSTM) and compared it with state-of-the-art STMs.

cs.DC↗

arXiv2019

An Efficient Approach to Achieve Compositionality using Optimized Multi-Version Object Based Transactional Systems

In the modern era of multi-core systems, the main aim is to utilize the cores properly. This utilization can be done by concurrent programming. But developing a flawless and well-organized concurrent program is difficult. Software Transactional Memory Systems (STMs) are a convenient programming interface which assist the programmer to access the shared memory concurrently without worrying about consistency issues such as priority-inversion, deadlock, livelock, etc. Another important feature that STMs facilitate is compositionality of concurrent programs with great ease. It composes different concurrent operations in a single atomic unit by encapsulating them in a transaction. Many STMs available in the literature execute read/write primitive operations on memory buffers. We represent them as Read-Write STMs or RWSTMs. Whereas, there exist some STMs (transactional boosting and its variants) which work on higher level operations such as insert, delete, lookup, etc. on a hash-table. We refer these STMs as Object Based STMs or OSTMs. The literature of databases and RWSTMs say that maintaining multiple versions ensures greater concurrency. This motivates us to maintain multiple version at higher level with object semantics and achieves greater concurrency. So, this paper pro-poses the notion of Optimized Multi-version Object Based STMs or OPT-MVOSTMs which encapsulates the idea of multiple versions in OSTMs to harness the greater concurrency efficiently.

cs.DC↗

arXiv2019

Achieving Starvation-Freedom in Multi-Version Transactional Memory Systems

Software Transactional Memory systems (STMs) have garnered significant interest as an elegant alternative for addressing synchronization and concurrency issues with multi-threaded programming in multi-core systems. Client programs use STMs by issuing transactions. STM ensures that transaction either commits or aborts. A transaction aborted due to conflicts is typically re-issued with the expectation that it will complete successfully in a subsequent incarnation. However, many existing STMs fail to provide starvation freedom, i.e., in these systems, it is possible that concurrency conflicts may prevent an incarnated transaction from committing. To overcome this limitation, we systematically derive a novel starvation free algorithm for multi-version STM. Our algorithm can be used either with the case where the number of versions is unbounded and garbage collection is used or where only the latest K versions are maintained, KSFTM. We have demonstrated that our proposed algorithm performs better than existing state-of-the-art STMs.

cs.DC↗

arXiv2018

CausalSpartanX: Causal Consistency and Non-Blocking Read-Only Transactions

Causal consistency is an intermediate consistency model that can be achieved together with high availability and performance requirements even in presence of network partitions. In the context of partitioned data stores, it has been shown that implicit dependency tracking using timestamps is more efficient than explicit dependency tracking. Existing time-based solutions depend on monotonic psychical clocks that are closely synchronized. These requirements make current protocols vulnerable to clock anomalies. In this paper, we propose a new time-based algorithm, CausalSpartanX, that instead of physical clocks, utilizes Hybrid Logical Clocks (HLCs). We show that using HLCs, without any overhead, we make the system robust on physical clock anomalies. This improvement is more significant in the context of query amplification, where a single query results in multiple GET/PUT operations. We also show that CausalSpartanX decreases the visibility latency for a given data item compared with existing time-based approaches. In turn, this reduces the completion time of collaborative applications where two clients accessing two different replicas edit same items of the data store. CausalSpartanX also provides causally consistent distributed read-only transactions. CausalSpartanX read-only transactions are non-blocking and require only one round of communication between the client and the servers. Also, the slowdowns of partitions that are unrelated to a transaction do not affect the performance of the transaction. Like previous protocols, CausalSpartanX assumes that a given client does not access more than one replica. We show that in presence of network partitions, this assumption (made in several other works) is essential if one were to provide causal consistency as well as immediate availability to local updates.

cs.DC↗

arXiv2018

An Innovative Approach to Achieve Compositionality Efficiently using Multi-Version Object Based Transactional Systems

In the modern era of multicore processors, utilizing cores is a tedious job. Synchronization and communication among processors involve high cost. Software transaction memory systems (STMs) addresses this issues and provide better concurrency in which programmer need not have to worry about consistency issues. Another advantage of STMs is that they facilitate compositionality of concurrent programs with great ease. Different concurrent operations that need to be composed to form a single atomic unit is achieved by encapsulating them in a single transaction. In this paper, we introduce a new STM system as multi-version object based STM (MVOSTM) which is the combination of both of these ideas for harnessing greater concurrency in STMs. As the name suggests MVOSTM, works on a higher level and maintains multiple versions corresponding to each key. We have developed MVOSTM with the unlimited number of versions corresponding to each key. In addition to that, we have developed garbage collection for MVOSTM (MVOSTM-GC) to delete unwanted versions corresponding to the keys to reduce traversal overhead. MVOSTM provides greater concurrency while reducing the number of aborts and it ensures compositionality by making the transactions atomic. Here, we have used MVOSTM for the list and hash-table data structure as list-MVOSTM and HT- MVOSTM. Experimental results of list-MVOSTM outperform almost two to twenty fold speedup than existing state-of-the-art list based STMs (Trans-list, Boosting-list, NOrec-list, list-MVTO, and list-OSTM). HT-MVOSTM shows a significant performance gain of almost two to nineteen times better than existing state-of-the-art hash-table based STMs (ESTM, RWSTMs, HT-MVTO, and HT-OSTM). MVOSTM with list and hash-table shows the least number of aborts among all the existing STM algorithms. MVOSTM satisfies correctness-criteria as opacity.

cs.DC↗

arXiv2018

Technical Report: Optimistic Execution in Key-Value Store

Limitations of the CAP theorem imply that if availability is desired in the presence of network partitions, one must sacrifice sequential consistency, a consistency model that is more natural for system design. We focus on the problem of what a designer should do if he/she has an algorithm that works correctly with sequential consistency but is faced with an underlying key-value store that provides a weaker (e.g., eventual or causal) consistency. We propose a detect-rollback based approach: The designer identifies a correctness predicate, say $P$, and continues to run the protocol, as our system monitors $P$. If $P$ is violated (because the underlying key-value store provides a weaker consistency), the system rolls back and resumes the computation at a state where $P$ holds. We evaluate this approach with practical graph applications running on the Voldemort key-value store. Our experiments with deployment on Amazon AWS EC2 instances shows that using eventual consistency with monitoring can provide a $50-80\%$ increase in throughput when compared with sequential consistency. We also show that the overhead of the monitoring itself is low (typically less than 4\%) and the latency of detecting violations is small. In particular, more than $99.9\%$ of violations are detected in less than $50$ milliseconds in regional AWS networks, and in less than $5$ seconds in global AWS networks.

cs.DC↗

arXiv2018

Toward Adaptive Causal Consistency for Replicated Data Stores

Causal consistency for key-value stores has two main requirements (1) do not make a version visible if some of its dependencies are invisible as it may violate causal consistency in the future and (2) make a version visible as soon as possible so that clients have the most recent information (to the extent feasible). These two requirements conflict with each other. Existing key-value stores that provide causal consistency (or detection of causal violation) utilize a static approach in the trade-off between these requirements. Depending upon the choice, it assists some applications and penalizes some applications. We propose an alternative where the system provides a set of tracking groups and checking groups. This allows the application to choose the settings that are most suitable for that application. Furthermore, these groups can be dynamically changed based on application requirements.

cs.DC↗

arXiv2018

Automated Synthesis of Distributed Self-Stabilizing Protocols

In this paper, we introduce an SMT-based method that automatically synthesizes a distributed self-stabilizing protocol from a given high-level specification and network topology. Unlike existing approaches, where synthesis algorithms require the explicit description of the set of legitimate states, our technique only needs the temporal behavior of the protocol. We extend our approach to synthesize ideal-stabilizing protocols, where every state is legitimate. We also extend our technique to synthesize monotonic-stabilizing protocols, where during recovery, each process can execute an most once one action. Our proposed methods are fully implemented and we report successful synthesis of well-known protocols such as Dijkstra's token ring, a self-stabilizing version of Raymond's mutual exclusion algorithm, ideal-stabilizing leader election and local mutual exclusion, as well as monotonic-stabilizing maximal independent set and distributed Grundy coloring.

cs.SE↗

arXiv2018

Optimistic Execution in Key-Value Store

Limitations of CAP theorem imply that if availability is desired in the presence of network partitions, one must sacrifice sequential consistency, a consistency model that is more natural for system design. We focus on the problem of what a designer should do if she has an algorithm that works correctly with sequential consistency but is faced with an underlying key-value store that provides a weaker (e.g., eventual or causal) consistency. We propose a detect-rollback based approach: The designer identifies a correctness predicate, say P , and continue to run the protocol, as our system monitors P . If P is violated (because the underlying key-value store provides a weaker consistency), the system rolls back and resumes the computation at a state where P holds. We evaluate this approach in the Voldemort key-value store. Our experiments with deployment of Voldemort on Amazon AWS shows that using eventual consistency with monitoring can provide 20 - 40% increase in throughput when compared with sequential consistency. We also show that the overhead of the monitor itself is small (typically less than 8%) and the latency of detecting violations is very low. For example, more than 99.9% violations are detected in less than 1 second.

cs.DC↗

arXiv2018

DKVF: A Framework for Rapid Prototyping and Evaluating Distributed Key-value Stores

We present our framework DKVF that enables one to quickly prototype and evaluate new protocols for key-value stores and compare them with existing protocols based on selected benchmarks. Due to limitations of CAP theorem, new protocols must be developed that achieve the desired trade-off between consistency and availability for the given application at hand. Hence, both academic and industrial communities focus on developing new protocols that identify a different (and hopefully better in one or more aspect) point on this trade-off curve. While these protocols are often based on a simple intuition, evaluating them to ensure that they indeed provide increased availability, consistency, or performance is a tedious task. Our framework, DKVF, enables one to quickly prototype a new protocol as well as identify how it performs compared to existing protocols for pre-specified benchmarks. Our framework relies on YCSB (Yahoo! Cloud Servicing Benchmark) for benchmarking. We demonstrate DKVF by implementing four existing protocols --eventual consistency, COPS, GentleRain and CausalSpartan-- with it. We compare the performance of these protocols against different loading conditions. We find that the performance is similar to our implementation of these protocols from scratch. And, the comparison of these protocols is consistent with what has been reported in the literature. Moreover, implementation of these protocols was much more natural as we only needed to translate the pseudocode into Java (and add the necessary error handling). Hence, it was possible to achieve this in just 1-2 days per protocol. Finally, our framework is extensible. It is possible to replace individual components in the framework (e.g., the storage component).

cs.DC↗

arXiv2017

Monitoring Partially Synchronous Distributed Systems using SMT Solvers

In this paper, we discuss the feasibility of monitoring partially synchronous distributed systems to detect latent bugs, i.e., errors caused by concurrency and race conditions among concurrent processes. We present a monitoring framework where we model both system constraints and latent bugs as Satisfiability Modulo Theories (SMT) formulas, and we detect the presence of latent bugs using an SMT solver. We demonstrate the feasibility of our framework using both synthetic applications where latent bugs occur at any time with random probability and an application involving exclusive access to a shared resource with a subtle timing bug. We illustrate how the time required for verification is affected by parameters such as communication frequency, latency, and clock skew. Our results show that our framework can be used for real-life applications, and because our framework uses SMT solvers, the range of appropriate applications will increase as these solvers become more efficient over time.

cs.DC↗