Proxy Caching Research Articles

A grid site reimagined: Building a fully cloud-native ATLAS Tier 2 on Kubernetes

The University of Victoria (UVic) operates an Infrastructure-asa-Service scientific cloud for Canadian researchers, and a Tier 2 site for the ATLAS experiment at CERN as part of the Worldwide LHC Computing Grid (WLCG). At first, these were two distinctly separate systems, but over time we have taken steps to migrate the Tier 2 grid services to the cloud. This process has been significantly facilitated by basing our approach on Kubernetes, a versatile, robust, and very widely adopted automation platform for orchestrating containerized applications. Previous work exploited the batch capabilities of Kubernetes to run grid computing jobs and replace the conventional grid computing elements by interfacing with the Harvester workload management system of the ATLAS experiment. However, the required functionality of a Tier 2 site encompasses more than just batch computing. Likewise, the capabilities of Kubernetes extend far beyond running batch jobs, and include for example scheduling recurring tasks and hosting long-running externally-accessible services in a resilient way. We are now undertaking the more complex and challenging endeavour of adapting and migrating all remaining services of the Tier 2 site — such as APEL accounting and Squid caching proxies, and in particular the grid storage element — to cloud-native deployments on Kubernetes. We aim to enable fully comprehensive deployment of a complete ATLAS Tier 2 site on a Kubernetes cluster via Helm charts, which will benefit the community by providing a streamlined and replicable way to install and configure an ATLAS site. We also describe our experience running a high-performance self-managed Kubernetes ATLAS Tier 2 cluster at the scale of 8 000 CPU cores for the last two years, and compare with the conventional setup of grid services.

METHODS FOR OPTIMIZING THE LOADING AND UPDATING OF WEB PAGES USING CLOUD TECHNOLOGIES

An essential component of modern systems built on cloud technologies, which determines the performance and efficiency of system operation, is caching methods and technologies. For systems that provide static information to end-users, technologies such as SSG (Static Site Generation) and SSR (Server-side Rendering) are applied in building the client side. SSG is typically used for pages with a very low frequency of content changes, as modifying the content on one page requires rebuilding the entire site. To cache SSR application pages, the use of a cache proxy server is recommended. An example of implementing such an approach is Next.js Serverless. However, because the Next.js Serverless technology employs a weak caching model, delivering up-to-date data to end users often takes a significant amount of time, which can negatively impact user interaction. This work is dedicated to addressing this problem. The paper proposes a flexible and efficient solution for generating and caching static pages using a cache proxy server and invalidation for data updates by the cache proxy server upon each change in the system's state. The integration of this solution into complex information systems is explored, and a performance comparison and evaluation of data delivery to end-users are conducted. The results of the conducted experiment demonstrate that the proposed approach solves the identified problem without compromising the performance compared to the Next.js Serverless approach.

Development of a method for processing log files using clustering

A log file is a document that keeps track of all events that occur on a website or server. Many log files are very large, so they can be regularly written over outdated content, or entire collections of log files with names, including a date, for example, can be created. In the event of technical problems, site inaccessibility, virus infection, hacker attacks and Distributed Denial of Service (DDoS) attacks, the resource administrator can use the information in log to find the cause, which makes it easier and faster to eliminate unwanted incidents. The paper analyzes the definition, types, location, use and examples of log files. Data are transferred to the MySQL database using the Squid.db database. Clustering is performed using a database. The study highlights clustering, analyzes metrics, and determines the proximity of clusters and objects in clusters in Euclidean space. Experiments are conducted and the results are satisfactory. For example, data are transferred to the MySQL database using the Squid.db database. Since the Squid proxy server is a cache proxy server, it stores resources, and the work is done quickly on the next request. Data are clustered using a compiled table of databases transferred to MySQL via Squid proxy. In this case, unnecessary entries are deleted from the table, which significantly speeds up data processing. The application of clustering method in problem solving is fast and simple. For the problem stated, the degree of closeness of clusters and objects in clusters in Euclidean space is determined. Experiments are conducted using obtained results.

Lightweight Robust Size Aware Cache Management

Modern key-value stores, object stores, Internet proxy caches, and Content Delivery Networks (CDN) often manage objects of diverse sizes, e.g., blobs, video files of different lengths, images with varying resolutions, and small documents. In such workloads, size-aware cache policies outperform size-oblivious algorithms. Unfortunately, existing size-aware algorithms tend to be overly complicated and computationally expensive. Our work follows a more approachable pattern; we extend the prevalent (size-oblivious) TinyLFU cache admission policy to handle variable-sized items. Implementing our approach inside two popular caching libraries only requires minor changes. We show that our algorithms yield competitive or better hit-ratios and byte hit-ratios compared to the state-of-the-art size-aware algorithms such as AdaptSize, LHD, LRB, and GDSF. Further, a runtime comparison indicates that our implementation is faster by up to 3× compared to the best alternative, i.e., it imposes a much lower CPU overhead.

Understanding the Practices of Global Censorship through Accurate, End-to-End Measurements

It is challenging to conduct a large scale Internet censorship measurement, as it involves triggering censors through artificial requests and identifying abnormalities from corresponding responses. Due to the lack of ground truth on the expected responses from legitimate services, previous studies typically require a heavy, unscalable manual inspection to identify false positives while still leaving false negatives undetected. In this paper, we propose Disguiser, a novel framework that enables end-to-end measurement to accurately detect the censorship activities and reveal the censor deployment without manual efforts. The core of Disguiser is a control server that replies with a static payload to provide the ground truth of server responses. As such, we send requests from various types of vantage points across the world to our control server, and the censorship activities can be recognized if a vantage point receives a different response. In particular, we design and conduct a cache test to pre-exclude the vantage points that could be interfered by cache proxies along the network path. Then we perform application traceroute towards our control server to explore censors' behaviors and their deployment. With Disguiser, we conduct 58 million measurements from vantage points in 177 countries. We observe 292 thousand censorship activities that block DNS, HTTP, or HTTPS requests inside 122 countries, achieving a 106 false positive rate and zero false negative rate. Furthermore, Disguiser reveals the censor deployment in 13 countries.

Understanding the Practices of Global Censorship through Accurate, End-to-End Measurements

It is challenging to conduct a large scale Internet censorship measurement, as it involves triggering censors through artificial requests and identifying abnormalities from corresponding responses. Due to the lack of ground truth on the expected responses from legitimate services, previous studies typically require a heavy, unscalable manual inspection to identify false positives while still leaving false negatives undetected. In this paper, we propose Disguiser, a novel framework that enables end-to-end measurement to accurately detect the censorship activities and reveal the censor deployment without manual efforts. The core of Disguiser is a control server that replies with a static payload to provide the ground truth of server responses. As such, we send requests from various types of vantage points across the world to our control server, and the censorship activities can be recognized if a vantage point receives a different response. In particular, we design and conduct a cache test to pre-exclude the vantage points that could be interfered by cache proxies along the network path. Then we perform application traceroute towards our control server to explore censors' behaviors and their deployment. With Disguiser, we conduct 58 million measurements from vantage points in 177 countries. We observe 292 thousand censorship activities that block DNS, HTTP, or HTTPS requests inside 122 countries, achieving a 10^-6 false positive rate and zero false negative rate. Furthermore, Disguiser reveals the censor deployment in 13 countries.

Analysis and visualization of proxy caching using LRU, AVL tree and BST with supervised machine learning

Proxy cache access speeds and decreases load time. There is an uncertainty regarding cache level 2 which you neglect. This research shall investigate the L1 cache, primary cache, and L2 cache as a secondary proxy server cache. LRU is typically utilised instead of cache. LRU for cold cache removal used to be a time-consuming process, but it isn't particularly efficient today. The performance of the cache L1 used LRU and L2 used LRU_AVL has risen with these solutions. The output is the LRU_AVL among other ways that utilize LRU tables and graphs. proxy cache LRU, LRU_AVL, and LRU_BST have average time of access calculated. Median access time is estimated using Python tools including Pandas, MatPlotLib as well as LRU, LRU_AVL, and LRU_BST. This research will anticipate the average period of usage of LRU, LRU_AVL, and LRU_BST cache algorithms.

Design of a Novel Architecture for Cost-Effective Cloud-based Content Delivery Network

Content Delivery Network (CDN) offers faster transmission of massive content from content providers to users using servers that are distributed geographically to offer seamless relay of service. However, conventional CDN is not capable of catering to the larger scope of demand for data delivery, and hence cloud-based CDN evolves as a solution. In a real-world scenario, each requested content has different popularity for different users. The problem arises with deciding which content objects should be placed in each content server to minimize delivery delays and storage costs. A review of existing approaches in cloud-based CDN shows that yet the problem associated with content placement is not solved. In this regard, a precise strategy is required to select the contents objects to be placed in a content server to achieve higher efficiency without affecting the CCDN performance. Therefore, the proposed system introduces a novel architecture that addresses this practical problem of content placement. The study considers placement problem as optimization problem with the ultimate purpose of maximizing the user content requests served and reducing the overall cost associated with content and data delivery. With an inclusion of a bucket-based concept for cache proxy and content provider, a novel topology is constructed where an optimal algorithm for placement of content is implemented using matrix operation of row reduction and column reduction. Simulation outcome shows that the proposed system excels better performance in contrast to the existing content placement strategy for cloud-based CDN.

Boosting Performance of Data-intensive Analysis Workflows with Distributed Coordinated Caching

Data-intensive end-user analyses in high energy physics require high data throughput to reach short turnaround cycles. This leads to enormous challenges for storage and network infrastructure, especially when facing the tremendously increasing amount of data to be processed during High-Luminosity LHC runs. Including opportunistic resources with volatile storage systems into the traditional HEP computing facilities makes this situation more complex.Bringing data close to the computing units is a promising approach to solve throughput limitations and improve the overall performance. We focus on coordinated distributed caching by coordinating workows to the most suitable hosts in terms of cached files. This allows optimizing overall processing efficiency of data-intensive workows and efficiently use limited cache volume by reducing replication of data on distributed caches.We developed a NaviX coordination service at KIT that realizes coordinated distributed caching using XRootD cache proxy server infrastructure and HTCondor batch system. In this paper, we present the experience gained in operating coordinated distributed caches on cloud and HPC resources. Furthermore, we show benchmarks of a dedicated high throughput cluster, the Throughput-Optimized Analysis-System (TOpAS), which is based on the above-mentioned concept.

Consistent hashing with bounded loads and virtual nodes-based load balancing strategy for proxy cache cluster

Response time and bandwidth occupancy of backbone network are greatly affected by load balancing and cache hit ratio in proxy cache cluster (PCC). A novel load balancing strategy is proposed to improve the homogeneous and heterogeneous PCC performance. By combining with the different performance parameters, performance ratio can be formulated with independent information data fluctuation weighting method, and the virtual nodes (VNs) attached to every cache node (CN) can be generated by a random function. The greatest common divisor m of the number of VNs attached to every CN can be calculated and the hash ring is cut into m arcs, and the VNs are mapped proportionally to arcs of the hash ring by the MD5 function. Requests are assigned to VNs based on the forwarding rule of consistent hashing with bounded loads, and the CN in correspondence with the VNs will be selected to provide services. Simulation results show that the strategy can significantly improve the load balancing in homogeneous and heterogeneous PCC, effectively reduce bandwidth occupancy of backbone network and PCC response time.

Algoritme penggantian cache proxy terdistribusi untuk meningkatkan kinerja server web

The performance of web processing needs to increase to meet the growth of internet usage, one of which is by using cache on the web proxy server. This study examines the implementation of the proxy cache replacement algorithm to increase cache hits in the proxy server. The study was conducted by creating a clustered or distributed web server system using eight web server nodes. The system was able to provide increased latency by 90 % better and increased throughput of 5.33 times better.

Web Proxy Auto Discovery for Dynamically Created Web Proxies

The WLCG Web Proxy Auto Discovery (WPAD) service provides a convenient mechanism for jobs running anywhere on the WLCG to dynamically discover web proxy cache servers that are nearby. The web proxy caches are general purpose for a number of different http applications, but different applications have different usage characteristics and not all proxy caches are engineered to work with the heaviest loads. For this reason, the initial sources of information for WLCG WPAD were the static configurations that ATLAS and CMS maintain for the Conditions data that they read through the Frontier Distributed Database system, which is the most demanding popular WLCG application for web proxy caches. That works well for use at traditional statically defined WLCG sites, but now that usage of commercial clouds is increasing, there is also a need for web proxy caches to dynamically register themselves as they are created. A package called Shoal had already been created to manage dynamically created web proxy caches. This paper describes the integration of the Shoal package into the WLCG WPAD system, such that both statically and dynamically created web proxy caches can be located from a single source. It also describes other improvements to the WLCG WPAD system since the last CHEP publication.

Evaluation of the ATLAS model for remote access to database resident information for LHC Run 3

The ATLAS model for remote access to database resident information relies upon a limited set of dedicated and distributed Oracle database repositories complemented with the deployment of Frontier system infrastructure on the WLCG (Worldwide LHC Computing Grid). ATLAS clients with network access can get the database information they need dynamically by submitting requests to a Squid proxy cache server in the Frontier network which provides results from its cache or passes new requests along the network to launchpads co-located at one of the Oracle sites (the master Oracle database at CERN or one of the Tier 1 Oracle database replicas). Since the beginning of LHC Run 1, the system has evolved in terms of client, Squid, and launchpad optimizations but the distribution model has remained fundamentally unchanged. On the whole, the system has been broadly successful in providing data to clients with relatively few disruptions even while site databases were down due to overall redundancy. At the same time, its quantitative performance characteristics, such as the global throughput of the system, the load distribution between sites, and the constituent interactions that make up the whole, were largely unknown. But more recently, information has been collected from launchpad and Squid logs into an Elasticsearch repository which has enabled a wide variety of studies of various aspects of the system. This contribution*** will describe dedicated studies of the data collected in Elasticsearch over the previous year to evaluate the efficacy of the distribution model. Specifically, we will quantify any advantages that the redundancy of the system offers as well as related aspects such as the geographical dependence of wait times seen by clients in getting a response to its requests. These studies are essential so that during LS2 (the long shutdown between LHC Run 2 and Run 3), we can adapt the system in preparation for the expected increase in the system load in the ramp up to Run 3 operations.

QoS3: Secure Caching in HTTPS Based on Fine-Grained Trust Delegation

With the ever-increasing concern in network security and privacy, a major portion of Internet traffic is encrypted now. Recent research shows that more than 70% of Internet content is transmitted using HyperText Transfer Protocol Secure (HTTPS). However, HTTPS encryption eliminates the advantages of many intermediate services like the caching proxy, which can significantly degrade the performance of web content delivery. We argue that these restrictions lead to the need for other mechanisms to access sites quickly and safely. In this paper, we introduce QoS3, which is a protocol that can overcome such limitations by allowing clients to explicitly and securely re-introduce in-network caching proxies using fine-grained trust delegation without compromising the integrity of the HTTPS content and modifying the format of Transport Layer Security (TLS). In QoS3, we classify web page contents into two types: (1) public contents that are common for all users, which can be stored in the caching proxies, and (2) private contents that are specific for each user. Correspondingly, QoS3 establishes two separate TLS connections between the client and the web server for them. Specifically, for private contents, QoS3 just leverages the original HTTPS protocol to deliver them, without involving any middlebox. For public contents, QoS3 allows clients to delegate trust to specific caching proxy along the path, thereby allowing the clients to use the cached contents in the caching proxy via a delegated HTTPS connection. Meanwhile, to prevent Man-in-the-Middle (MitM) attacks on public contents, QoS3 validates the public contents by employing Document object Model (DoM) object-level checksums, which are delivered through the original HTTPS connection. We implement a prototype of QoS3 and evaluate its performance in our testbed. Experimental results show that QoS3 provides acceleration on page load time ranging between 30% and 64% over traditional HTTPS with negligible overhead. Moreover, QoS3 is deployable since it requires just minor software modifications to the server, client, and the middlebox.

A Framework to improve the Web Performance using Reorganization, Optimized Prediction and Prefetching

Improving web performance is becoming more hectic in recent days. This paperelucidates the combination of many ideas to improve web performance and given as a framework. The entire framework depicts various aspects in improving web access performance which includes website reorganization, webpage prediction and prefetching, optimized way of accessing prediction algorithm in webserver and finally improvements in a proxy cache at the time of accessing dynamic content. Each portion of the framework has been successfully proposed and implemented. The various algorithms have been introduced in each portion of the implementation. This research work not only introduced new algorithms but also create scope for further research works in terms of improving web performance.

Performance Analysis of Multithreaded IoT Gateway

This paper presents a constrained application protocol enabled multithreaded Internet of Things (IoT) gateway that allows reliable and quick access to big data from IoT domains. We consider a proxy caching policy to accelerate the content delivery rate of sensor data while reducing the energy consumption of corresponding nodes. We propose a selective caching policy in which resource validation requests are sent based on user requests, besides updates from IoT nodes. We also describe an analytical model to evaluate the performance of multithreaded IoT gateway under various traffic conditions to meet the quality of service constraints. The proposed model is evaluated over single hop IPv6 over low-power wireless personal area networks/IEEE 802.15.4 cluster. The performance evaluation clearly shows the effectiveness of our model in providing fast response and reduced energy consumption of the resources.

Accelerating the Internet in the presence of Big Data: Reducing user delays by leveraging historical user request patterns for web caching

Approximately 4 billion people have access to the Internet, additionally 23 billion devices are connected as of 2018. This has allowed for a substantial growth in data collection which has allowed for Big Data to flourish. The continued increase in user, devices, and Big Data usage has created a significant intensification in Internet traffic. This in turn has the potential to increase user delays when accessing data on the Internet. There are a number of ways to help reduce user latency, web caching is able to reduce web user delays in addition to reducing network traffic and the load on web servers. In this study we propose a proxy level web caching mechanism leveraging historical web patterns to help reduce user latency and accelerate the Internet. In addition we survey the state of the art of other caching approaches. Our investigation shows that using historical patterns as part of a proxy caching mechanisms in large scale networks can significantly shorten the latency for users in this era of Big Data

Enhancing Greedy Web Proxy caching using Weighted Random Indexing based Data Mining Classifier

Web Proxy caching system is an intermediary between the Web users and servers that try to alleviate the loads on the origin servers by caching particular Web objects and behaves as the proxy for the server and services the requests that are made to the servers. In this paper, the performance of a Proxy system is measured by the number of hits at the Proxy. Higher number of hits at the Proxy server reflects the effectiveness of the Proxy system. The number of hits is determined by the replacement policies chosen by the Proxy systems. Traditional replacement policies that are based on time and size are reactive and do not consider the events that will possibly happen in the future. The performance of the web proxy caching system is improved by adapting Data Mining Classifier model based on Web User clustering and Weighted Random Indexing Methods. The outcome of the paper are proactive strategies that augment the traditional replacement policies such as GDS, GDSF, GD∗ which uses the Data Mining techniques.

The Open High Throughput Computing Content Delivery Network

LHC experiments make extensive use of web proxy caches, especially for software distribution via the CernVM File System and for conditions data via the Frontier Distributed Database Caching system. Since many jobs read the same data, cache hit rates are high and hence most of the traffic flows efficiently over Local Area Networks. However, it is not always possible to have local web caches, particularly for opportunistic cases where experiments have little control over site services. The Open High Throughput Computing (HTC) Content Delivery Network (CDN), openhtc.io, aims to address this by using web proxy caches from a commercial CDN provider. Cloudflare provides a simple interface for registering DNS aliases of any web server and does reverse proxy web caching on those aliases. The openhtc.io domain is hosted on Cloudflare's free tier CDN which has no bandwidth limit and makes use of data centers throughout the world, so the average performance for clients is much improved compared to reading from CERN or a Tier 1. The load on WLCG servers is also significantly reduced. WLCG Web Proxy Auto Discovery is used to select local web caches when they are available and otherwise select openhtc.io caching. This paper describes the Open HTC CDN in detail and provides initial results from its use for LHC@Home and USCMS opportunistic computing.

PCR: caching replacement algorithm in proxy server

The efficiency of caching is a key factor affecting the performance of content delivery network (CDN). The main aim of current CDN caching is to obtain a higher hit ratio, since the validation and freshness of outdated pages have not received due consideration in their replacement model. In this paper, a new improved cache profit model is clearly defined, and the freshness factors of web pages have been taken into adequate account. Based on the profit model, a new replacement algorithm – proxy cache replacement (PCR) – is recommended, and it can be proved optimal under the rational hypothesis. To conclude, a series of comparative experiments verified the efficiency of PCR in web caching replacement.