DOCSLIB.ORG

Smart Allocation and Replication of Memory for Parallel Programs

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Smart Allocation and Replication of Memory for Parallel Programs Shoal: Smart Allocation and Replication of Memory For Parallel Programs Stefan Kaestle, Reto Achermann, and Timothy Roscoe, ETH Zürich; Tim Harris, Oracle Labs, Cambridge https://www.usenix.org/conference/atc15/technical-session/presentation/kaestle This paper is included in the Proceedings of the 2015 USENIX Annual Technical Conference (USENIC ATC ’15). July 8–10, 2015 • Santa Clara, CA, USA ISBN 978-1-931971-225 Open access to the Proceedings of the 2015 USENIX Annual Technical Conference (USENIX ATC ’15) is sponsored by USENIX. Shoal: smart allocation and replication of memory for parallel programs Stefan Kaestle, Reto Achermann, Timothy Roscoe, Tim Harris∗ Systems Group, Dept. of Computer Science, ETH Zurich ∗Oracle Labs, Cambridge, UK Abstract mization to apply in which situation and, with rapidly Modern NUMA multi-core machines exhibit complex la- evolving and diversifying hardware, programmers must tency and throughput characteristics, making it hard to repeatedly make manual changes to their software to allocate memory optimally for a given program’s access keep up with new hardware performance properties. patterns. However, sub-optimal allocation can signifi- One solution to achieve better data placement and cantly impact performance of parallel programs. faster data access is to rely on automatic online moni- We present an array abstraction that allows data place- toring of program performance to decide how to migrate ment to be automatically inferred from program analysis, data [13]. However, monitoring may be expensive due to and implement the abstraction in Shoal, a runtime library missing hardware support (if pages must be unmapped for parallel programs on NUMA machines. In Shoal, to trigger a fault when data is accessed) or insufficiently arrays can be automatically replicated, distributed, or precise (if based on sampling using performance coun- partitioned across NUMA domains based on annotating ters). Both approaches are limited to a relatively small memory allocation statements to indicate access patterns. number of optimizations (e.g. it is hard to incrementally We further show how such annotations can be auto- activate large pages or switch to using DMA hardware matically provided by compilers for high-level domain- for data copies based on monitoring or event counters) specific languages (for example, the Green-Marl graph We present Shoal, a system that abstracts memory ac- language). Finally, we show how Shoal can exploit ad- cess and provides a rich programming interface that ac- ditional hardware such as programmable DMA copy en- cepts hints on memory access patterns at the runtime. gines to further improve parallel program performance. These hints can either be manually written or automati- We demonstrate significant performance benefits from cally derived from high-level descriptions of parallel pro- automatically selecting a good array implementation grams such as domain specific languages. Shoal includes based on memory access patterns and machine charac- a machine-aware runtime that selects optimal implemen- teristics. We present two case-studies: (i) Green-Marl, tations for this memory abstraction dynamically during a graph analytics workload using automatically anno- buffer allocation based on the hints and a concrete com- tated code based on information extracted from the high- bination of machine and workload. If available, Shoal level program and (ii) a manually-annotated version of is able to exploit not only NUMA properties but also the PARSEC Streamcluster benchmark. hardware features such as large pages and DMA copy engines. Our contributions are: 1 Introduction a memory abstraction based on arrays that decou- • Memory allocation in NUMA multi-core machines is ples data access from the rest of the program, an interface for programs to specify memory access increasingly complex. Good placement of and access • to program data is crucial for application performance, patterns when allocating memory, a runtime that selects from several highly tuned ar- and, if not carefully done, can significantly impact scal- • ability [3, 13]. Although there is research (e.g. [7, 3]) ray implementations based on access patterns and in adapting to the concrete characteristics of such ma- machine characteristics and can exploit machine chines, many programmers struggle to develop software specific hardware, features modifications to Green-Marl [20], a graph analyt- applying these techniques. We show an example in Sec- • tion 5.1. ics language, to show how Shoal can extract access The problem is that it is unclear which NUMA opti- patterns automatically from high-level descriptions. USENIX Association 2015 USENIX Annual Technical Conference 263 Node 2 Node 0 GDDR RAM RAM memory allocation strategy. Memory is not allocated di- Memory DMA DMA MC MC DMA rectly when calling malloc, but mapped only when the L3 Cache L3 Cache corresponding memory is first accessed by a thread. This IC resulting page fault will cause Linux to back the faulting Xeon Phi page from the NUMA node of the faulting core. IC IC A surprising consequence of this choice is that on GPGPU GPGPU Linux the implementation of the initialization phase of DMA IC a program is often critical to its memory performance, L3 Cache L3 Cache Node 3 Node 1 even through programmers rarely consider initialization DMA MC MC DMA GDDR as a candidate for heavy optimization, since it almost Memory RAM RAM never dominates the total execution time of the program. MC Memory Controllers IC Interconnect Link 16x PCI Express Link Figure 1: Architecture of a modern multi-core machine. To see why, consider that memset is the most widely used approach for initializing the elements of an array. Most programmers will spend little time evaluating al- 2 Motivation ternatives, since the time spent in the initialization phase Modern multi-core machines have complex memory is usually negligible. An example is as follows: hierarchies consisting of several memory controllers // ---- Initialization (sequential) ----------- placed across the machine – for example, Figure 1 shows void *ptr = malloc(ARRSIZE); such a machine with four host main memory controllers memset(ptr, 0, ARRSIZE); (one per processor socket). Memory latency and band- // ---- Work (parallel, highly optimized) ----- width depend on which core accesses which memory lo- execute_work_in_parallel(); cation [8, 10]. The interconnect may suffer congestion The scalability of a program written this way can be when access to memory controllers is unbalanced [13]. limited. memset executes on a single core and so all Future machines will be more complex: they may not memory is allocated on the NUMA node of that core. provide global cache coherence [21, 25], or even shared For memory-bound parallel programs, one memory con- global physical addresses [1]. Even today, accelerators troller will be saturated quickly while others remain idle like Intel’s Xeon Phi, GPGPUs, and FPGAs have higher since all threads (up to 64 on the machines we evaluate) memory access costs for parts of the physical memory request memory from the same controller. Furthermore, space [1]. Such hardware demands even more care in the interconnect close to this memory controller will be application data placement. more susceptible to congestion. This poses challenges to programmers when allocat- There are two problems here: (i) memory is not allo- ing and accessing memory. First, detailed knowledge cated (or mapped) when the interface suggests (memory of hardware characteristics and a good understanding of is not allocated inside malloc itself but later in the exe- their implications for algorithm performance is needed cution) and (ii) the choice of where to allocate memory is for efficient scalable programs. Care must be taken when made in a subsystem (the OS kernel) that has no knowl- choosing a memory controller to allocate memory from, edge of the intended access patterns of this memory. and how to subsequently access that memory. This can be addressed by tuning algorithms to spe- Second, hardware changes quickly meaning that de- cific operating systems. For example, we could initialize sign choices must be constantly re-evaluated to ensure memory using a parallel for loop: good performance on current hardware. This imposes high engineering and maintenance costs. This is worth- // ---- Initialization (parallel) ------------- while in high-performance computing or niche markets, void *ptr = malloc(ARRSIZE); but general purpose machines have too broad a hardware #pragma omp parallel for range for this to be practical for many domains. The re- for (int i=0;i<ARRSIZE; i++) sult is poor performance on most platforms. init(i); // ---- Work (parallel, highly optimized) ----- These problems can be seen in much code today. Pro- execute_work_in_parallel(); grammers take little care of where memory is allocated and how it is accessed. In cases like the popular Stream- This will be faster and retain scalability in current cluster benchmark (evaluated in Section 5.1) and appli- versions of Linux. The first-touch strategy will equally cations from the NAS benchmark suite [13], memory is spread out memory across all memory controllers, which allocated using a low-level malloc call which provides balances the load on them and reduces contention on in- no guarantees about where memory is allocated or other dividual interconnect links. details such as the page size to use. One drawback of this strategy is the loss of portabil- For example, Linux currently employs a first-touch ity and scalability when the OS kernel’s internal memory 2 264 2015 USENIX Annual Technical Conference USENIX Association allocation policies change. Furthermore, it also requires high-level constructs in the PageRank example can be correct setup of OpenMP’s CPU affinity to ensure that all determined as follows: (i) Foreach (T: G.Nodes) cores participate in this parallel initialization phase in or- means the nodes-array will be accessed sequen- der to spread memory equally on all memory controllers. tially, read-only, and with an index, and (ii) Sum(w: Finally, we might do better: allocate memory close to the t.InNbrs) implies read-only, indexed accesses on in- cores that access it the most.
Load more

Recommended publications
  • From Francois De Quesnay Le Despotisme De La Chine (1767)
    From Francois de Quesnay Le despotisme de la Chine (1767) [translation by Lewis A. Maverick in China, A Model for Europe (1946)] [Francois de Quesnay (1694-1774) came from a modest provincial family but rose into the middle class with his marriage. He was first apprenticed to an engraver but then turned to the study of surgery and was practicing as a surgeon in the early 1730's. In 1744 he received a medical degree and was therefore qualified as a physician as well as a surgeon and shifted his activity in the latter direction. In 1748, on the recommendation of aristocratic patients, he became physician to Madame de Pompadour, Louis XV's mistress. In 1752 he successfully treated the Dauphin (crown prince) for smallpox and was then appointed as one of the physicians to the King. While serving at court he became interested in economic, political, and social problems and by 1758 was a member of the small group of economists known as "Physiocrats." Their main contribution to economic thinking involved analogizing the economy to the human body and seeing economic prosperity as a matter of ensuring the circulation of money and goods. The Physiocrats became well-known throughout Europe, and were visited by both David Hume and Adam Smith in the 1760s. The Physiocrats took up the cause of agricultural reform, and then broadened their concern to fiscal matters and finally to political reform more generally. They were not democrats; rather they looked to a strong monarchy to adopt and push through the needed reforms. Like many other Frenchman of the mid 18th-century, the Physiocrats saw China as an ideal and modeled many of their proposals after what they believed was Chinese practice.
    [Show full text]
  • Bicycle Manual Road Bike
    PURE CYCLING MANUAL ROAD BIKE 1 13 14 2 3 15 4 a 16 c 17 e b 5 18 6 19 7 d 20 8 21 22 23 24 9 25 10 11 12 26 Your bicycle and this manual comply with the safety requirements of the EN ISO standard 4210-2. Important! Assembly instructions in the Quick Start Guide supplied with the road bike. The Quick Start Guide is also available on our website www.canyon.com Read pages 2 to 10 of this manual before your first ride. Perform the functional check on pages 11 and 12 of this manual before every ride! TABLE OF CONTENTS COMPONENTS 2 General notes on this manual 67 Checking and readjusting 4 Intended use 67 Checking the brake system 8 Before your first ride 67 Vertical adjustment of the brake pads 11 Before every ride 68 Readjusting and synchronising 1 Frame: 13 Stem 13 Notes on the assembly from the BikeGuard 69 Hydraulic disc brakes a Top tube 14 Handlebars 16 Packing your Canyon road bike 69 Brakes – how they work and what to do b Down tube 15 Brake/shift lever 17 How to use quick-releases and thru axles about wear c Seat tube 16 Headset 17 How to securely mount the wheel with 70 Adjusting the brake lever reach d Chainstay 17 Fork quick-releases 71 Checking and readjusting e Rear stay 18 Front brake 19 How to securely mount the wheel with 73 The gears 19 Brake rotor thru axles 74 The gears – How they work and how to use 2 Saddle 20 Drop-out 20 What to bear in mind when adding them 3 Seat post components or making changes 76 Checking and readjusting the gears 76 Rear derailleur 4 Seat post clamp Wheel: 21 Special characteristics of carbon 77
    [Show full text]
  • This World Is but a Canvas to Our Imagination
    PARENT NOTES Geography / Art Understanding Chines / Creative Response At A Glance: About this activity The post-visit task links to specific skills in Geography and Art that ü Geography / Art Crossover will be engaging and relevant to students in key stage 1. ü Key Stage 1 (ideally Y2) Following their visit to Blackgand Chine, children will learn about the basics of coastal erosion. They will learn how a chine is formed ü National Curriculum 2014 and understand that erosion can demolish chines completely. They will investigate some of the features of chines and understand ü Practises investigation and representation skills that they are diferent because of the process of erosion. They will recognise some similarities and diferences. ü Follow-up activity Following this, they produce a piece of art work which is based on images of chines. Their final task will be to locate chines on a map What’s Involved? of the Isle of Wight. Following their visit to Blackgand Questions & Answers Chine, children will learn about the What is the task? basics of coastal erosion.They will This is an art task which uses a geographical stimulus. investigate some of the features of chines and understand that they are Afer studying a number of photographs of island chines, diferent because of the process of children will choose an image which they wish to re-create, erosion. using a variety of media. They will learn how to create texture within a piece of art work and use this to good efect. Following this, they produce a piece Once all of the pieces of art have been completed, students of art work which is based on images will gather together and see how many can be matched to the of chines.
    [Show full text]
  • Constructing Reservoir Dams in Deglacierizing Regions of the Nepalese Himalaya the Geneva Challenge 2018
    Constructing reservoir dams in deglacierizing regions of the Nepalese Himalaya The Geneva Challenge 2018 Submitted by: Dinesh Acharya, Paribesh Pradhan, Prabhat Joshi 2 Authors’ Note: This proposal is submitted to the Geneva Challenge 2018 by Master’s students from ETH Zürich, Switzerland. All photographs in this proposal are taken by Paribesh Pradhan in the Mount Everest region (also known as the Khumbu region), Dudh Koshi basin of Nepal. The description of the photos used in this proposal are as follows: Photo Information: 1. Cover page Dig Tsho Glacial Lake (4364 m.asl), Nepal 2. Executive summary, pp. 3 Ama Dablam and Thamserku mountain range, Nepal 3. Introduction, pp. 8 Khumbu Glacier (4900 m.asl), Mt. Everest Region, Nepal 4. Problem statement, pp. 11 A local Sherpa Yak herder near Dig Tsho Glacial Lake, Nepal 5. Proposed methodology, pp. 14 Khumbu Glacier (4900 m.asl), Mt. Everest valley, Nepal 6. The pilot project proposal, pp. 20 Dig Tsho Glacial Lake (4364 m.asl), Nepal 7. Expected output and outcomes, pp. 26 Imja Tsho Glacial Lake (5010 m.asl), Nepal 8. Conclusions, pp. 31 Thukla Pass or Dughla Pass (4572 m.asl), Nepal 9. Bibliography, pp. 33 Imja valley (4900 m.asl), Nepal [Word count: 7876] Executive Summary Climate change is one of the greatest challenges of our time. The heating of the oceans, sea level rise, ocean acidification and coral bleaching, shrinking of ice sheets, declining Arctic sea ice, glacier retreat in high mountains, changing snow cover and recurrent extreme events are all indicators of climate change caused by anthropogenic greenhouse gas effect.
    [Show full text]
  • Defence Appraisal
    Directorate of Economy & Environment Director Stuart Love Isle of Wight Shoreline Management Plan 2 Appendix C: Baseline Process Understanding C2: Defence Appraisal December 2010 Coastal Management; Directorate of Economy & Environment, Isle of Wight Council iwight.com Appendix C2: Page 1 www.coastalwight.gov.uk/smp iwight.com Appendix C2: Page 2 www.coastalwight.gov.uk/smp Appendix C: Baseline Process Understanding C2: Defence Appraisal Contents C2.1 Introduction and Methodology 5 Acknowledgements 1. Introduction 5 2. Step 1: Residual Life based on Condition Grade 5 2.1 Data availability 2.2 Method 2.2.1 Condition 2.2.2 Residual Life 2.3 Results 2.3.1 Referencing of the Defences 2.3.2 Defence Appraisal Spreadsheet 2.3.3 GIS Mapping 2.4 Developing the ‘No Active Intervention’ Scenario 2.5 Developing the ‘With Present Management’ Scenario 2.6 Discussion 3. Step 2: Approval by the Defence Asset Managers 11 4. Coastal Defence Strategies 11 5. References: List of data sources used to inform the defence appraisal 12 C2.2 Defence Appraisal Tables 13 Including: Map showing the units used in the table 14 C2.3 Defence Appraisal Summary Maps iwight.com Appendix C2: Page 3 www.coastalwight.gov.uk/smp Acknowledgements Defra SMP Guidance (2006) was used in the production of this Defence Appraisal in 2009: Tasks 1.5c and 2.1b require the collation and assessment of necessary information on existing defences in accordance with Volume 2 & Appendix D, section 2.3b. Key data sources used in the production of this report: • Isle of Wight Council asset records, fully updated for this report; • NFCDD records (available only for Estuaries) Acknowledgement is given to the assistance provided by the Environment Agency Area Office and through Environment Agency support through Mott Macdonald.
    [Show full text]
  • Cumulative Watershed Effects of Fuel Management in the Western United States Elliot, William J.; Miller, Ina Sue; Audin, Lisa
    United States Department of Agriculture Forest Service Rocky Mountain Research Station General Technical Report RMRS-GTR-231 January 2010 Cumulative Watershed Effects of Fuel Management in the Western United States Elliot, William J.; Miller, Ina Sue; Audin, Lisa. Eds. 2010. Cumulative watershed effects of fuel management in the western United States. Gen. Tech. Rep. RMRS-GTR-231. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 299 p. ABSTRACT Fire suppression in the last century has resulted in forests with excessive amounts of biomass, leading to more severe wildfires, covering greater areas, requiring more resources for suppression and mitigation, and causing increased onsite and offsite damage to forests and watersheds. Forest managers are now attempting to reduce this accumulated biomass by thinning, prescribed fire, and other management activities. These activities will impact watershed health, particularly as larger areas are treated and treatment activities become more widespread in space and in time. Management needs, laws, social pressures, and legal findings have underscored a need to synthesize what we know about the cumulative watershed effects of fuel management activities. To meet this need, a workshop was held in Provo, Utah, on April, 2005, with 45 scientists and watershed managers from throughout the United States. At that meeting, it was decided that two syntheses on the cumulative watershed effects of fuel management would be developed, one for the eastern United States, and one for the western United States. For the western synthesis, 14 chapters were defined covering fire and forests, machinery, erosion processes, water yield and quality, soil and riparian impacts, aquatic and landscape effects, and predictive tools and procedures.
    [Show full text]
  • A Modelling Applied to Restoration Scenarios
    Article The interplay of flow processes shapes aquatic invertebrate successions in floodplain channels - A modelling applied to restoration scenarios MARLE, Pierre, et al. Abstract The high biotic diversity supported byfloodplains is ruled by the interplay of geomorphic and hydrological pro-cesses at various time scales, from dailyfluctuations to decennial successions. Because understanding such pro-cesses is a key question in river restoration, we attempted to model changes in taxonomic richness in anassemblage of 58 macroinvertebrate taxa (21 gastropoda and 37 ephemeroptera, plecoptera and trichoptera,EPT) along two successional sequences typical for former braided channels. Individual models relating the occur-rence of taxa to overflow and backflow durations were developed fromfield measurements in 19floodplainchannels of the Rhônefloodplain (France) monitored over 10 years. The models were combined to simulate di-versity changes along a progressive alluviation and disconnection sequence after the reconnection with the mainriver of a previously isolated channel. Two scenarios were considered: (i) an upstream + downstream reconnec-tion creating a lotic channel, (ii) a downstream reconnection creating a semi-lotic channel. Reconnection led to adirect increase in invertebrate richness (on average [...] Reference MARLE, Pierre, et al. The interplay of flow processes shapes aquatic invertebrate successions in floodplain channels - A modelling applied to restoration scenarios. Science of the Total Environment, 2021, vol. 750, no. 142081
    [Show full text]
  • TS14K Streambank Armor Protection with Stone Structures
    Technical Streambank Armor Protection with Supplement 14K Stone Structures (210–VI–NEH, August 2007) Technical Supplement 14K Streambank Armor Protection with Part 654 Stone Structures National Engineering Handbook Issued August 2007 Cover photo: Interlocking stone structures may be needed to provide a stable streambank. Advisory Note Techniques and approaches contained in this handbook are not all-inclusive, nor universally applicable. Designing stream restorations requires appropriate training and experience, especially to identify conditions where various approaches, tools, and techniques are most applicable, as well as their limitations for design. Note also that prod- uct names are included only to show type and availability and do not constitute endorsement for their specific use. (210–VI–NEH, August 2007) Technical Streambank Armor Protection with Supplement 14K Stone Structures Contents Purpose TS14K–1 Introduction TS14K–1 Benefits of using stone TS14K–1 Stone considerations TS14K–1 Design considerations TS14K–3 Placement of rock TS14K–5 Dumped rock riprap ......................................................................................TS14K–5 Machine-placed riprap ...................................................................................TS14K–6 Treatment of high banks TS14K–8 Embankment bench method ........................................................................TS14K–8 Excavated bench method .............................................................................TS14K–8 Surface flow protection TS14K–8
    [Show full text]
  • Random River: Trade and Rent Extraction in Imperial China
    Random river: Trade and rent extraction in imperial China Marlon Seror * December 2020 Abstract This paper exploits changes in the course of the Yellow River in China to isolate exogenous variation in the natural distribution of economic centers across space and over 2,000 years. Using original data on population and taxation from dynastic histories and local gazetteers, I assess the effect of market access on population density and resource extraction. I find that the changes in connectedness have two opposite effects. First, they induce a very large increase in the level and concentration of economic activity in the short run. Second, they trigger a large increase in taxation per capita and an elite flight, which reverse the concentration effect in the longer run. *Universit´edu Qu´ebec `aMontr´eal,IRD{DIAL, [email protected]. I thank James Fenske, Pei Gao, Stephan Heblich, Cl´ement Imbert, Ruixue Jia, Christian Lamouroux, Yu-Hsiang Lei, David Serfass, Lingwei Wu, Yu Zheng, Yanos Zylberberg, and seminar participants in Bristol and at the 2019 India-China Conference in Warwick for helpful comments and suggestions. The usual disclaimer applies. 1 Introduction As the economy grows, economic activity becomes increasingly concentrated in few cities. This process can be thought of as a consequence of economies of scale and ag- glomeration economies, and thus as conducive to economic development (Kim, 1995; Henderson et al., 2001). However, our understanding of agglomeration economies may be limited by the literature's focus on short-term economic effects, keeping po- litical structures constant. This paper uses data covering 2,000 years to study the very long run, as political structures react.
    [Show full text]
  • Reverse-Time Modeling of Channelized Meandering Systems from Geological Observations Marion Parquer
    Reverse-time modeling of channelized meandering systems from geological observations Marion Parquer To cite this version: Marion Parquer. Reverse-time modeling of channelized meandering systems from geological observa- tions. Applied geology. Université de Lorraine, 2018. English. NNT : 2018LORR0081. tel-01902547 HAL Id: tel-01902547 https://hal.univ-lorraine.fr/tel-01902547 Submitted on 5 Apr 2019 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. AVERTISSEMENT Ce document est le fruit d'un long travail approuvé par le jury de soutenance et mis à disposition de l'ensemble de la communauté universitaire élargie. Il est soumis à la propriété intellectuelle de l'auteur. Ceci implique une obligation de citation et de référencement lors de l’utilisation de ce document. D'autre part, toute contrefaçon, plagiat, reproduction illicite encourt une poursuite pénale. Contact : [email protected] LIENS Code de la Propriété Intellectuelle. articles L 122. 4 Code de la Propriété Intellectuelle. articles L 335.2- L 335.10 http://www.cfcopies.com/V2/leg/leg_droi.php http://www.culture.gouv.fr/culture/infos-pratiques/droits/protection.htm Reverse-time modeling of channelized meandering systems from geological observations Modélisation rétro-chronologique de systèmes chenalisés méandriformes à partir d’observations géologiques T HÈSE présentée et soutenue publiquement le 5 avril 2018 pour l’obtention du grade de Docteur de l’Université de Lorraine Spécialité Géosciences par Marion PARQUER Composition du jury: Rapporteurs: Prof.
    [Show full text]
  • Report IOW 5: Chilton Chine to Colwell Chine
    www.gov.uk/englandcoastpath England Coast Path Stretch: Isle of Wight Report IOW 5: Chilton Chine to Colwell Chine Part 5.1: Introduction Start Point: Chilton Chine (grid reference 440896.257, 82191.428) End Point: Colwell Chine (grid reference 432773.445, 87932.217) Relevant Maps: IOW 5a to IOW 5j 5.1.1 This is one of a series of linked but legally separate reports published by Natural England under section 51 of the National Parks and Access to the Countryside Act 1949, which make proposals to the Secretary of State for improved public access along and to the Isle of Wight coast. 5.1.2 This report covers length IOW 5 of the stretch, which is the coast between Chilton Chine to Colwell Chine. It makes free-standing statutory proposals for this part of the stretch, and seeks approval for them by the Secretary of State in their own right under section 52 of the National Parks and Access to the Countryside Act 1949. 5.1.3 The report explains how we propose to implement the England Coast Path (“the trail”) on this part of the stretch, and details the likely consequences in terms of the wider ‘Coastal Margin’ that will be created if our proposals are approved by the Secretary of State. Our report also sets out: any proposals we think are necessary for restricting or excluding coastal access rights to address particular issues, in line with the powers in the legislation; and any proposed powers for the trail to be capable of being relocated on particular sections (“roll- back”), if this proves necessary in the future because of coastal change.
    [Show full text]
  • Palaeoenvironment and Taphonomy of Dinosaur Tracks in the Vectis Formation (Lower Cretaceous) of the Wessex Sub-Basin, Southern England
    Cretaceous Research (1998) 19, 471±487 Article No. cr970107 Palaeoenvironment and taphonomy of dinosaur tracks in the Vectis Formation (Lower Cretaceous) of the Wessex Sub-basin, southern England *Jonathan D. Radley1, *Michael J. Barker and {Ian C. Harding * Department of Geology, University of Portsmouth, Burnaby Building, Burnaby Road, Portsmouth, PO1 3QL, UK; 1current address: Postgraduate Research Institute for Sedimentology, The University of Reading, PO Box 227, Whiteknights, Reading RG6 6AB, UK { Department of Geology, University of Southampton, Southampton Oceanography Centre, European Way, Southampton, SO14 3ZH, UK Revised manuscript accepted 20 June 1997 Reptilian ichnofossils are documented from three levels within the coastal lagoonal Vectis Formation (Wealden Group, Lower Cretaceous) of the Wessex Sub-basin, southern England (coastal exposures of the Isle of Wight). Footprints attributable to Iguanodon occur in arenaceous, strongly trampled, marginal lagoonal deposits at the base of the formation, indicating relatively intense ornithopod activity. These were rapidly buried by in¯uxes of terrestrial and lagoonal sediment. Poorly-preserved footcasts within the upper part of the Barnes High Sandstone Member are tentatively interpreted as undertracks. In the stratigraphically higher Shepherd's Chine Member, footcasts of a small to med- ium-sized theropod and a small ornithopod originally constituted two or more trackways and are pre- served beneath a distinctive, laterally persistent bioclastic limestone bed, characterised by hypichnial Diplocraterion. These suggest relatively low rates of dinosaurian activity on a low salinity, periodically wetted mud¯at. Trackway preservation in this case is due to storm-induced shoreward water move- ments which generated in¯uxes of distinctive bioclastic lithologies from marginal and offshore lagoo- nal settings.
    [Show full text]