Self Organizing Organic Fault Tolerance Computer Science Essay

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Abstract: Business applications for the web need to be fault tolerant. Having human monitoring of such large scale applications and reacting to their internal faults is not always possible. Intelligent Agent-based systems can provide self-healing capabilities and tolerance to faults by dynamically configuring paths in real-time. In this paper, we present a self-healing organic computing based multi-agent framework for reduction of down time in an enterprise-level distributed system. We develop extensive agent-based simulations using cellular automata to validate the framework. Our results demonstrate an overall significant reduction in downtime.

Keywords: Agent-based simulation, fault-tolerance, self-healing, self-organization, peer-to-peer, wireless sensor network, mobile, dependable, validation, organic computing, cellular automata.

1. Introduction

Modern Distributed Enterprise Systems are typically transactional in nature. Users access the systems from some kind of front end, which can range from a web application to even voice based menus. However, behind the user interface, lies the magic of a large-scale distributed system, which typically ends in one or more data-centre(s). This system is required to be operational 24/7, 365 days an year. However, it is not humanly possible to keep monitoring all the paths and internal details of this system. Thus fault-tolerance needs to be a property of the components of the system. In this paper, we address this aspect of enterprise business systems by developing an agent-based self-healing fault tolerance scheme, which automatically adapts to faults. We develop an agent-based simulation to validate the framework.

The rest of the paper is structured as following:

First we present background and related work. Subsequently, we discuss in detail the ideas and develop a thorough theoretical framework of fault-tolerance for SOFT, Organic Self-Organizing Fault Tolerance. Next, we present the case study and the development of an agent-based simulation model. We then discuss in detail the results by comparing them with those of traditional systems without self-organization. Finally, we conclude the paper.

2. Background and Related Work

2.1 The Organic Computing: Engineered Self-Organization

Large-scale peer-to-peer and wireless network systems occasionally require Self-* properties. Organic Computing encompasses these self-organizing, self-healing and in-general self-* properties from the natural world, which in turn, can be applied to technical systems. The idea behind Organic computing is to develop efficient adaptive systems in which a balance is to be developed between flexibility, stability and robustness. Such systems are typically specialized as they need to focus on solving instances of certain problem classes such as the ones faced in their environments [1].

Basic concepts of Organic Computing are discussed in [2] and [3]. [4] discusses the feasibility of controlled emergence using organic computing. [5] addresses controlled self-organization to address complexity. [6] talks about using Organic Computing as a vision of self-organising systems adapting robustly to dynamically changing environments without running out of control. It focuses on the design of a generic system architecture, which allows for self-organisation but at the same time enables adequate reactions to control the sometimes completely unexpected emerging global behaviour of these self-organised technical systems. [1] discusses the coupling of evolution and learning in organic computing systems. It uses nervous systems as examples. The paper discusses trends and recent results in combining evolutionary and neural computation are reviewed. The emphasis is put on the influence of evolution and development on the structure of neural systems. It demonstrates how neural structures can be evolved that efficiently learn solutions for problems from a particular problem class. Simple examples of systems that "learn to learn" as well as technical solutions for the design of turbo-machinery components are presented. [7] discusses the use of organic computing in distributed embedded systems. [8] discusses Self-Organizing Computation on the Organic Grid.

2.2 Self-Healing Systems

In this section, we discuss background and theory of self-healing systems.

A survey of self-healing systems is presented in [9]. [10] discusses performance evaluation for self-healing distributed services and fault-detection mechanisms. Fast self-healing gradients are discussed in [11]. Self-healing components in robust software architecture for concurrent and distributed systems are discussed in [12]. [13] discusses a self-healing and scalable fault-tolerant network for runtime environments. It is implemented on top of Scalable fault-tolerant protocol (SFTP). [14] talks about heterogeneous self-organizing nodes for establishing trust relationships and uses real-life negotiations as well as dependability. [15] discusses the future of microcomputers such as using spray computers and self-organization. [16] introduces fault recovery techniques useful in the design of re-configurable holonic manufacturing systems.

2.3 Distributed Enterprise Transactional Systems

In this section, we discuss the peculiarities of distributed enterprise transactional systems.

2.4 Handling Cascading Faults

In this section, we discuss the problem of cascading faults. We first discuss cascading faults and fault tolerance in other literature and then present the case of the need for fault tolerance against them.

Although not directly related, Self-Checking networks using cascades in logic gates are given in [17]. In [18], Julich et al. demonstrate a fault detection and redundancy management system for a dual redundancy based network architecture where the principal control components are configured and programmed to repitively carry out performance tests as an a priori requirement for network command capability. The domain of this work is however in Computer Architecture.

In [19], novel method for monitoring Industrial process plants such as chemical refineries and electric power generation is presented. This technique exploits the system's dynamic behaviour for diagnostic clues. The key techniques are: modelling the physical system with dynamic qualitative/quantitative models, inducing diagnostic knowledge from qualitative simulations, continuously comparing observations against fault-model predictions, and incrementally creating and testing multiple-fault hypotheses. The important result is that the diagnosis is refined as the physical system's dynamic behaviour is revealed over time. However, it does not handle cascading faults.

In [20], Pertet and Narasimham discuss emergent "vulnerability of complexity" which increases the likelihood of cascading failures that might result in widespread disruption. The paper explored whether and how to exploit the knowledge of the system's topology, the application's interconnections and the application's normal fault-free behavior to build proactive fault-tolerance techniques that could curb the spread of cascading failures and enable faster system-wide recovery. The paper also characterizes the topology knowledge entailment as well as the benefits and tradeoffs of this approach.

[21] presents a strategy for the prevention of cascading outages in power systems. The strategy employs wide-area backup protection and artificial intelligence techniques to minimize the impact of a fault on a network. Two by using wide-area backup protection for the prevention of cascading trips leading to a wide spread blackout are given. The first is based on accurately locating the fault. The second is the avoidance of unnecessary trips, due to hidden failure or overloading, by blocking the trip signals of conventional backup protection relays. In addition, the paper discusses the limitations of conventional backup protection are examined and presents some suggestions for improvement.

In the domain of business systems, cascading faults, unlike simple faults, occur in complex scenarios where failure or overload of single component, leads to a series of faults resulting in an overall poor customer service experience. The user is unaware of all this, except for the fact, the extremely long waiting time. The actual reason for the fault is the basic naïve fault-tolerant scheme typically used in distributed business systems. What happens is that the typical solution to a fault is to dynamically switch to an alternate pre-defined path because of an overload of a single path (e.g. on some special occasion, after Holidays etc.). Now, as all traffic switches to the alternate path, this path gets overloaded as well. This makes all the traffic switch to another path and so on. Thus, such faults can be extremely hard to monitor and a nightmare for the support staff to deal with.

3. SOFT: A Detailed Description

3.1 Basic Definitions

3.2 Formalization

In this section, we conceptualize intelligent self-adaptive nodes and develop a model. We evaluate strategies for the nodes to adapt based on the Hobbesian model.

3. Model Development

3.1 Fault Tolerance modelling using Cellular Automata

Cellular Automata are an excellent means of investigation of social phenomenon [22-29] as well as self-organization [30, 31]. With the prevalence of large scale of social networking systems, we demonstrate that the CA approach can also be very useful in the representation of models of links and exploring paradigms of fault-tolerance. In the following figure, we see the basic model for Cellular Automata. Here, the black cells (squares) represent the nodes, the grey ones represent the links in something similar to the Moore's neighbourhood form [32] so that each node has 8 links to its surrounding nodes. Since this is different from the original Moore's neighbourhood, we shall call it Modified Moore's neighbourhood to avoid confusion. Here faults are shown by red links. The basic idea is that each node needs to keep track of its working links so how much processing and communication power does it need to spend to maintain this information. This situation is important in the case of Wireless sensor nodes where nodes are having limited power. The goal is to have discover patterns so a single node can intelligently decide its fault-monitoring probability (e.g. via Sending ping messages or heartbeat or other communication strategy).

Fig. 1. Basic CA Model.

Fig. 2. Modified Moore's neighbourhood for peer node.

3.2 Differential equations Model

4. Simulation Experiments

4.1Experimental setup

The experimental setup consists of NetLogo Behavior Space experiments {NetLogo}, a tool widely used for agent-based modelling and simulation and especially cellular automata modelling {Nigel07}. Each point on the plot is generated by running 30 simulation experiments and is plotted to minimize with confidence intervals to cater for effects of random numbers. In each experiment, other than changing the primary variable, we also vary the run time from 100 to 1000 steps to ensure consistency.

4.2 Varying the Fault Time

In this experiment, we vary the fault rate from 5 to 20 faults per second. The probability of faults is constant at % and similarly self-healing probability is constant at %. The time for self-healing is .

4.3 Varying the Fault Probability

In this experiment, we vary the fault probability from 0 to 100 faults per second. The time for faults is constant at % and similarly self-healing probability is constant at %. The time for self-healing is .

5. Discussion of Results

3.1 Fault Tolerance

6. Conclusion & Future Directions

These settings

A research paper should be structured in terms of four parts logically, each of which may comprise multiple sections:

Part 1: is problem description/definition, and a literature review upon the state of the art;

Part 2: is methodological formulation and/or theoretical development (fundamentals, principle and/or approach, etc.);

Part 3: is prototyping, case study or experiment;

Part 4: is critical evaluation against related works, and the conclusion.

A survey paper may skip Part 3, but should multiply Part 2 and elaborate Parts 1 and 4.

An application paper may light touch Part 2 but should elaborate Part 3, with Parts 1 and 4 similar to what a research paper would.

In any article it is unnecessary to have an arrangement (i.e., housekeeping) statement at the beginning (or end) of every (Sub-)Section. Rather, a single brief statement about the arrangement of the whole paper can be made at the end of the Introduction Section.

SI units should be used, i.e., the units based on the meter, kilogram, second, etc.

2.3 Tables

Number tables as Table 1, Table 2 etc, and every table must be referred to in the body of text properly.

Each table should have a brief and self-explanatory title, within 1-2 lines.

Table's column headers should be short, but sufficiently explanatory. Standard abbreviations of units of measurement should be added between parentheses.

Vertical lines should not be used to separate columns. Leave some extra space between the columns instead.

The vertical frame lines of cells of table should be set zero width (i.e., no lines) where appropriate, and the leftmost and right most frame lines must be zero width. By this way, a table will look as Tables 1 and 2.

Table 1. Tabular example one.

Column head 1

Column head 2

Row 1

A, a

A, b

Row 2

B, a

B, b

Table 2. Tabular example two.

Column head 1

Column head 2

Row 1

A, a

A, b

Row 2

B, a

B, b

2.4 Figures

Number figures as Fig. 1, Fig. 2, etc., both in the captions of figures and in the citation of figures in the text, and every Figure must be referred to in the body of text, properly.

Each figure should have a self-explanatory caption, within 1-2 lines.

Any setting that prevents figure/graphics from being printed legibly on black-and-white printer should be avoided. This includes, e.g., color figures/graphics, color background/shading in figures, figures/graphics pasted from the Internet, display screen, video/image shots, etc.

For figures please take the following points into account:

in general, a figure/graphics should be generated in resolution of 1200 dpi;

line art should have a minimum resolution of 600 dpi;

grayscales (including photos) should have a minimum resolution of 300 dpi (no lettering), or 500 dpi (when there is lettering);

do not use figures pasted from the Internet or other display sources, the resolution will be too low for printing;

do not use color in your figures if they are to be printed in black and white, as this will reduce or simply black out the print quality (note that in software often the default is color, you should change the settings);

figures/graphics can be scaled down or up, but this must be with extra care as scaling may distort the original shape (visual trueness) of the figure/graphics;

figures should be designed with the format of the page of the journal in mind. They should be of such a size as to allow a reduction of 50%;

texts in figures/graphics must be sufficiently legible, with visually similar size as those in the body of text;

big figure/graphics can be placed across both columns of the layout.

On maps and other figures where a scale is needed, use bar scales rather than numerical ones, i.e., do not use scales of the type 1:10,000. By this way problems can be avoided if the figures need to be reduced.

Photographs, images, graphics, etc., are only acceptable if they have good contrast and intensity.

2.5 Self-Defined Headers

Self-defined headers are, for instance, Definition, Theorem, Proof, Algorithm, Process, etc. A header is followed by an Arabic numeral with dot (.). Text continues in the same line of header.

2.6 Equations

All equations should be numbered at right-hand end consecutively throughout body of text and appendices, respectively, and should be justified to the right-hand side but keeping the body of equation looking like being centered. Symbols and texts in equations should look like 9 - 10 points.

Table 3. Sizes of characters in equations.

Size

Full

10 point

Subscript/superscript

7 point

Sub- subscript/superscript

5 point

Symbols

10 point

Sub-symbols

7 point

An equation can be cited in the text in the form of the number in parentheses preceded with "Eq.", e.g., Eq. (3), Eq. (5), etc.

2.7 References and Citation in Text

All items of references should be placed after body of text.

Item of reference with missing filed(s), e.g., missing pagination, missing volume number, etc., is unacceptable.

Citation of references in text. Every item of reference must be cited in the body of text properly, in the form of a single numeral in square brackets, e.g., [6] [12] [21]. Citation of consecutive reference items may be shortened, e.g., [3] - [10].

2.8 Notes

Keep a Note as short as possible. Number notes, with not header, in 1, 2, etc., with dot (.), and place all notes together at the end of paper, following references.

When cited in the text, precede the number of the note with header Note and place them in parentheses, e.g., (Note 1), (Note 2), etc. Notes' citation should neither be superscript nor subscript.

2.9 Appendices

Numbered as Appendix A, Appendix B, etc., and cited in the body of text accordingly.

2.10 Author Bios (ITSSA Journal Only)

For articles of ITSSA Journal only, not for CoSIWN Journal.

An author bio should not exceed 20 lines of text, no photo.

2.11 Numbering Systems

(Sub-)sections, figures, tables, equations, etc.

Arabic numerals (e.g., 1, 2, 3 …; 2.1, 2.2 …)

Self-defined headers

Header (e.g., Definition, Theorem, Proof, Algorithm, Process, etc.) followed by Arabic numerals with dot (.))

Numbered listings

Lower-case roman sequence in ( ), e.g., (i), (ii), (iii); or lower-case alphabet in ( ), e.g., (a), (b), (c)

No numbering systems should be linked with sectional numbers, except in Appendices.

2.12 English

Use "United States" English.

Avoid use of first persons (i.e., "I", "we", "us", "our", etc.), which can nicely be expressed by the passive voice or other ways. This also applies to the Abstract.

For non-native English speaking authors,

try to keep sentences short and simple;

try to follow the basic English grammars;

try to keep the same word for the same thing or concept. Do not artificially diversify vocabulary;

seek the advice of a native English speaker, if possible, before submitting a manuscript.

3. Specifications of Camera Ready Version (CRV)

3.1 Default Settings of the Printing throughout a Paper

These settings apply to the whole paper, including body of text and all appended sections, see Tables 4 and 5 and Fig. 1.

Table 4. Layout.

Paper size

A4 (21 cm x 29.7 cm)

Fonts

10 points, Times New Rome

Single column

Title, author information (names, affiliation, postal address, Email, URL)

Double column,

with spacing between columns = 0.5 cm

Abstract, keywords, body of text, appended sections

Line spacing

Single-line spacing

Justification of texts

Fully justified, i.e., flush both left and right

Table 5. Margins.

All pages,

including title and subsequent pages

Top / Bottom margin

2.5 cm

Left / Right margin

1.5 cm

All margin area (top, bottom, left and right) should be kept blank, no pagination, no footnotes, no running title, or similar.

Use of grid on document will affect single-line spacing. Ensure document grid is disabled in the format settings.

3.2 Indentions and Blank Lines

No indention to the 1st line

to a paragraph immediately following a blank line

1st line indention by 0.4 cm

to a paragraph immediately following non-blank line

hanging by 0.6 cm (i.e., with the 2nd and subsequent lines indention)

to all listings (bulleted, numbered, or text-outlined),

1st, 2nd section headings,

Reference items,

Notes items

No blank lines between paragraphs, including before, between and after listings (bulleted, numbered, or text-outlined), and items of reference, and before or after equations.

All blank line should be 10 points (except the two 14 points blank lines preceding the title) and non boldface, regardless of the size and boldface of the text lines before or after.

Hanging should be adjusted to 0.8 cm for 3rd, 4th level section headings and where smaller hanging will leave raggedness.

2nd, and 3rd level indention should be the same hanging s the 1st level but will have 0.6 cm and 1.2 cm indention from left, respectively.

Avoid excessive spacing before or after section headings, figure, tables, etc. Two blank lines are only applicable in three places on the title page, i.e., before (in 14 points) and after the paper's title and before the Abstract, from which texts are formatted in double columns. In all other cases, only actually apply only one blank line.

3.3 Titles and Section Headings

Numbering

Font size

Boldface

Justified

Case

Before/After Spacing (all 10 points blank line except preceding the title)

Title

--

14 points

NONE

Centered

Title case

2 blank lines (14 points) before, and

2 blank lines (10 points) after

Author names

--

12 points

NONE

Centered

Title case

1 blank line after author name line

Author affiliation/address

--

10 points

NONE

Centered

Title case

2 blank lines after the very last line of address info

"Abstract" and "Keywords"

--

10 points

YES (Header only)

Flush left and right

NONE.

Header with colon (:) continued by texts

1st level section headings

Number with dot (.)

12 points

YES

Flush left and right

Title case

1 blank line before and after

1st level appended sections

NONE

12 points

YES

Flush left and right

Title case

1 blank line before and after

2nd and 3rd level section headings

YES

10 points

YES

Flush left and right

Title case

1 blank line before and after

4th and subsequent level section headings

NONE

10 points

NONE.

But underlined

Flush left and right

Sentence case

NONE.

Heading continued by texts

Figure captions

"Fig." plus number with dot (.)

10 points

NONE

Centered

Sentence case

1 blank line before and after.

Header continued by caption ended with dot (.)

Titles of tables

"Table" plus number with dot (.)

10 points

NONE

Aligned with table

Sentence case

1 blank line before and after.

Header continued by title ended with dot (.)

Self-defined headers

Header plus number with dot (.)

10 points

YES

(Header and number only)

Flush left and right

Sentence case

1 blank line before.

Header continued by texts ended with dot (.)

Title case: Capitalize the first letter of nouns, pronouns, verbs, adjectives, and adverbs; do not capitalize articles, coordinate conjunctions, or prepositions (unless the title or the headings begins with such a word).

Nicely hang up (i.e., indent the 2nd and subsequent lines of) 1st, 2nd and 3rd level section headings. Do not end with dot (.) any level section headings, though 1st level section heading number is with a dot (.).

Appended sections may be (in order): Acknowledgements, References, Notes, Appendices, Author Bios.

3.4 Author Names and Affiliations

For multiple authorships,

put all author's names in one line;

place superscript number after each author's surname;

begin affiliation/address of each author by the numeral;

avoid repetition if parts of affiliation/address/Email are the same.

By convention: first name comes before surname, no title. Use comma (,) or "and" (for the last pair) to separate multiple authors.

Keep each set of affiliation/address within 4 lines.

No blank line between different sets of affiliations/addresses. After all the affiliations/addresses, place 2 blank lines (in 10 points).

Email and URL should not be activated in hyperlink, nor underlined. "Email" is default word, no need to add "URL".

3.5 Abstract and Keywords

No line break or paragraph break (i.e., no hard return "ï‚¿") after the default word "Abstract"/"Keywords". Instead, insert a colon (:) after the headers, and texts continue in the same lines of headers.

No blank line between Abstract and Keywords.

Abstract should not exceed 20 lines of text.

Keywords should be arranged in an alphabetic order and should not exceed 8 keywords. Texts of keywords should be uniformly singular, lower case (except abbreviations), separated by comma (,), ended with a dot (.)

(a) Title page. (b) Subsequent pages.

Fig. 1. Illustrations of layouts.

3.5 Listings

Listings can be bulleted, but can also be numbered, or even text-outlined.

Nicely hang up (i.e., indent the 2nd and subsequent lines of) listings. See Sub-Section 3.2 about Indentions.

Reference items and Notes items are examples of numbered listings.

In text-outlined listings, the preceding texts can be headers (e.g., steps of algorithms or pseudo codes, etc.) followed by Arabic numerals with dot (.), or specific words, phrases or even sentences. the preceding texts of text-outlined listings should not be underlined, but can be set as Italic or Bold fonts (not the dot) or Title Case, followed by dot (.) if the preceding text is not part of the following text, or by semi-colon (;) if otherwise. For example,

-Operator. There are two types of operations, i.e., numerical and symbolic. -Operator is a symbolic mapping from Region A to B.

Region A: is the area represented in the robot by Map A.

3.6 Figures and Tables

Figure and caption

centered;

caption below figure

Table and title

title above table and aligned with table

A figure/table can occupy full single column, where necessary. Basically, texts in figures and tables should look like 9 - 10 points.

Figure and table should not stride on page break. Place figure and table properly amidst the body of texts such that no excessive blank is left at the bottom or top of a page.

Figure or table across both columns should be placed either at the top or the bottom of a page, by breaking text appropriately.

While ideally figure or table should be placed after their reference in the text as immediately as it is practically possible, it is acceptable figure or table is placed prior to or a little far from their reference in the text, if so required by the layout, so long as they are placed in the same order of their numbering sequence. However, in any case, avoid placing all figures and table at the end of the text in a lumping manner.

3.7 References

Items of references should be ordered and numbered according to the alphabetical order of author's surnames and the chronological order for same authors, with consecutive numbers in square brackets, e.g., [1] [2] etc.

Nicely hang up (i.e., indent the 2nd and subsequent lines of) each item of reference, after the ending square bracket ( ] ).

End each item with a dot (.) except for URL.

No blank lines between items of references.

Author's names should be Initials (without dots for omission) followed by Surnames; multiple author names are divided by comma (,) except for the last two authors by conjunction "and".

URL's. List URL all together and place them following other items of reference, de-activate hyperlink, no underline. Item of URL can be justified to the left, i.e., flush left.

The fields of a reference item should be arranged in the order and the cases as shown in Table 3.

Table 3. Order of the fields of a reference item.

(a) Journal items.

Authors (Initials + Surnames)

Title of article

Name of Journal

Vol. #

No. #

Month Year

pp. #

Case

Title Case

Sentence case

Title Case

Title Case

Title Case

Title Case

Lower case

Delimiter

Comma

Dot

Comma

Comma

Comma

Comma

(b) Conference items.

Authors (Initials + Surnames)

Title of article

Name of Conference (Acronym)

Place, Dates, Month, Year

Vol. #

pp. #

(# pages)

Case

Title Case

Sentence case

Title Case

Title Case

Title Case

Lower case

Delimiter

Comma

Dot

Comma

Comma

Comma

(c) Contribution in book items.

Authors (Initials + Surnames)

Title of article

Editors (Initials + Surnames)

Name of Book

Publisher, Place

Month Year

pp. #

Case

Title Case

Sentence case

Title Case

Title Case

Title Case

Title Case

Lower case

Delimiter

Comma

Dot

Comma

Comma

Comma

Comma

(d) Book items.

Authors/ Editors (Initials + Surnames)

Name of Book

Publisher, Place

Month Year

pp. #

Case

Title Case

Title Case

Title Case

Title Case

Lower case

Delimiter

Comma

Comma

Comma

Comma

3.8 Length Limits

CoSIWN Journal. Unless otherwise notified explicitly by the Editor, there is normally no length limit on an article for publication.

ITSSA. Normally there is no length limit on article for publication unless, in some circumstance, the length limit may be applied, in which case the Editor will notify you explicitly.

3.10 Electronic File

Authors must format their papers to comply with the specifications as described in this Instruction.

Once a paper is accepted, the Author should provide the following electronic files of their CRV, named by the paper ID and packed in a compressed file (or placed on a URL for temporary bespoken access):

the Word Doc (97 version) of the article (or the latex source files), with figures and tables properly embedded amidst the text;

a PDF file that is generated from (i). Texts must be preserved in the PDF. PDF printed as image is unacceptable;

separate PDF files of high-resolution, high-quality graphic files of all figures and tables, one PDF file per figure or table, named as fig01, fig02a, fig02b … The PDF file of each figure should be scaled to full paper size, with figure's captions being excluded/deleted and any blank edges all around being trimmed off;

plain text file of title, authors, abstract and keywords; and

scanned copy of signed copyright form.

Acknowledgements

This Instruction represents numerous useful comments and suggestions from authors, editors, and the editorial staff over many years.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.