BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20240310T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20231105T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20231221T175617Z
UID:0E356283-0C89-423F-ABC8-C1E28657B01C
DTSTART;TZID=America/New_York:20231218T144500
DTEND;TZID=America/New_York:20231218T163000
DESCRIPTION:We will examine the RowHammer problem in Dynamic Random Access 
 Memory (DRAM)\, the first example of how a circuit-level failure mechanism
  can cause a practical and widespread system security vulnerability. RowHa
 mmer is the phenomenon that repeatedly accessing a row in a modern DRAM ch
 ip predictably causes bitflips in physically-adjacent rows. Building on ou
 r initial fundamental work that appeared at ISCA 2014\, Google Project Zer
 o demonstrated that this hardware phenomenon can be exploited by user-leve
 l programs to gain kernel privileges. Many other works demonstrated other 
 attacks exploiting RowHammer\, including remote takeover of a server vulne
 rable to RowHammer\, takeover of a mobile device by a malicious user-level
  application\, and destruction of predictive capabilities of commonly-used
  deep neural networks.\n\nUnfortunately\, the RowHammer problem still plag
 ues cutting-edge DRAM chips\, DDR4 and beyond. Based on our recent charact
 erization studies of more than 1500 DRAM chips from six technology generat
 ions that appeared at ISCA 2020 and MICRO 2021\, we show that RowHammer at
  the circuit level is getting much worse\, newer DRAM chips are much more 
 vulnerable to RowHammer than older ones\, and existing mitigation techniqu
 es do not work well. We also show that existing proprietary mitigation tec
 hniques employed in DDR4 DRAM chips\, which are advertised to be Rowhammer
 -free\, can be bypassed via many-sided hammering (also known as TRRespass 
 &amp; Uncovering TRR).\n\nIn this talk\, we will provide an overview of RowHam
 mer research in academia and industry\, with a special focus on recent wor
 ks that rigorously analyze real chip characteristics and introduce promisi
 ng solution ideas. We will discuss the effect of RowHammer on High-Bandwid
 th Memory (HBM) chips and introduce and analyze RowPress\, which is a fund
 amentally different read disturbance phenomenon that also affects all DRAM
  chips. RowPress greatly (e.g.\, by 100X) reduces the activation count req
 uired to induce bitflips\, by keeping an activated row open for a long tim
 e. We will also discuss what other problems may be lurking in DRAM and oth
 er types of memory\, which can potentially threaten the foundations of rel
 iable and secure systems\, as memory technologies scale to higher densitie
 s. We will conclude by describing and advocating a principled approach to 
 memory robustness (including reliability\, security\, safety) research tha
 t can enable us to better anticipate and prevent such vulnerabilities.\n\n
 A short accompanying paper\, which appeared at ASP-DAC 2023\, can be found
  here and serves as recommended reading:\n&quot;Fundamentally Understanding and
  Solving RowHammer&quot;\nhttps://arxiv.org/abs/2211.07613\n\nCo-sponsored by: 
 IEEE North Jersey Section\n\nSpeaker(s): Onur Mutlu\n\nAgenda: \nEvent Tim
 e: 2:45 PM to 4:30 PM\n\nECE 202\, NJIT\, Newark\n\n3:45 PM to 3:00 PM Ref
 reshments and Networking\n\n3:00 PM to 4:30 PM Talk by Professor Onur Mutl
 u of ETH/Carnegie Mellon/Stanford\n\nSeminar is in ECEC 202. All Welcome: 
 There is no fee/charge for attending IEEE technical seminar. You don&#39;t hav
 e to be an IEEE Member to attend. Refreshments are free for all attendees.
  Please invite your friends and colleagues to take advantage of this talk.
 \n\nRoom: 202\, Bldg: Electrical and Computer Engineering\, 154 Summit Str
 eet\, Newark\, NJ 07102\, NJIT\, Newark\, New Jersey\, United States\, 071
 02
LOCATION:Room: 202\, Bldg: Electrical and Computer Engineering\, 154 Summit
  Street\, Newark\, NJ 07102\, NJIT\, Newark\, New Jersey\, United States\,
  07102
ORGANIZER:dmisra@njit.edu
SEQUENCE:59
SUMMARY:RowHammer\, RowPress and Beyond: Can We Be Free of Bitflips (Soon)?
URL;VALUE=URI:https://events.vtools.ieee.org/m/390503
X-ALT-DESC:Description: &lt;br /&gt;&lt;div dir=&quot;auto&quot;&gt;We will examine the RowHammer
  problem in Dynamic Random Access Memory (DRAM)\, the first example of how
  a circuit-level failure mechanism can cause a practical and widespread sy
 stem security vulnerability.&amp;nbsp\; RowHammer is the phenomenon that repea
 tedly accessing a row in a modern DRAM chip predictably causes bitflips in
  physically-adjacent rows. Building on our initial fundamental work that a
 ppeared at ISCA 2014\, Google Project Zero demonstrated that this hardware
  phenomenon can be exploited by user-level programs to gain kernel privile
 ges. Many other works demonstrated other attacks exploiting RowHammer\, in
 cluding remote takeover of a server vulnerable to RowHammer\, takeover of 
 a mobile device by a malicious user-level application\, and destruction of
  predictive capabilities of commonly-used deep neural networks.&lt;/div&gt;\n&lt;di
 v dir=&quot;auto&quot;&gt;&amp;nbsp\;&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;Unfortunately\, the RowHammer 
 problem still plagues cutting-edge DRAM chips\, DDR4 and beyond. Based on 
 our recent characterization studies of more than 1500 DRAM chips from six 
 technology generations that appeared at ISCA 2020 and MICRO 2021\, we show
  that RowHammer&amp;nbsp\; at the circuit level is getting much worse\, newer 
 DRAM chips are much more vulnerable to RowHammer than older ones\, and exi
 sting mitigation techniques do not work well. We also show that existing p
 roprietary mitigation techniques employed in DDR4 DRAM chips\, which are a
 dvertised to be Rowhammer-free\, can be bypassed via many-sided hammering 
 (also known as TRRespass &amp;amp\; Uncovering TRR).&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;&amp;n
 bsp\;&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;In this talk\, we will provide an overview of
  RowHammer research in academia and industry\, with a special focus on rec
 ent works that rigorously analyze real chip characteristics and introduce 
 promising solution ideas. We will discuss the effect of RowHammer on High-
 Bandwidth Memory (HBM) chips and introduce and analyze RowPress\, which is
  a fundamentally different read disturbance phenomenon that also affects a
 ll DRAM chips. RowPress greatly (e.g.\, by 100X) reduces the activation co
 unt required to induce bitflips\, by keeping an activated row open for a l
 ong time. We will also discuss what other problems may be lurking in DRAM 
 and other types of memory\, which can potentially threaten the foundations
  of reliable and secure systems\, as memory technologies scale to higher d
 ensities. We will conclude by describing and advocating a principled appro
 ach to memory robustness (including reliability\, security\, safety) resea
 rch that can enable us to better anticipate and prevent such vulnerabiliti
 es.&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;&amp;nbsp\;&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;A short accompan
 ying paper\, which appeared at ASP-DAC 2023\, can be found here and serves
  as recommended reading:&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;&quot;Fundamentally Understandi
 ng and Solving RowHammer&quot;&lt;/div&gt;\n&lt;div dir=&quot;auto&quot;&gt;&lt;a href=&quot;https://arxiv.or
 g/abs/2211.07613&quot; target=&quot;_blank&quot; rel=&quot;noopener noreferrer&quot; data-saferedir
 ecturl=&quot;https://www.google.com/url?q=https://arxiv.org/abs/2211.07613&amp;amp\
 ;source=gmail&amp;amp\;ust=1702521679697000&amp;amp\;usg=AOvVaw3kbEUzgr7tOOw0G1nci
 b34&quot;&gt;https://arxiv.org/abs/2211.&lt;wbr /&gt;07613&lt;/a&gt;&lt;/div&gt;&lt;br /&gt;&lt;br /&gt;Agenda: 
 &lt;br /&gt;&lt;p&gt;Event Time: 2:45 PM to 4:30 PM&lt;/p&gt;\n&lt;p&gt;ECE 202\, NJIT\, Newark&lt;/p
 &gt;\n&lt;p&gt;&lt;span style=&quot;font-family: -apple-system\, BlinkMacSystemFont\, &#39;Sego
 e UI&#39;\, Roboto\, Oxygen\, Ubuntu\, Cantarell\, &#39;Open Sans&#39;\, &#39;Helvetica Ne
 ue&#39;\, sans-serif\;&quot;&gt;3:45 PM to 3:00 PM Refreshments and Networking&lt;/span&gt;&lt;
 /p&gt;\n&lt;p&gt;&lt;span style=&quot;font-family: -apple-system\, BlinkMacSystemFont\, &#39;Se
 goe UI&#39;\, Roboto\, Oxygen\, Ubuntu\, Cantarell\, &#39;Open Sans&#39;\, &#39;Helvetica 
 Neue&#39;\, sans-serif\;&quot;&gt;3:00 PM to 4:30 PM Talk by &lt;strong&gt;Professor Onur Mu
 tlu of ETH/Carnegie Mellon/Stanford&lt;/strong&gt;&lt;/span&gt;&lt;/p&gt;\n&lt;p&gt;&lt;span style=&quot;f
 ont-family: -apple-system\, BlinkMacSystemFont\, &#39;Segoe UI&#39;\, Roboto\, Oxy
 gen\, Ubuntu\, Cantarell\, &#39;Open Sans&#39;\, &#39;Helvetica Neue&#39;\, sans-serif\;&quot;&gt;
 Seminar is in ECEC 202. All Welcome: There is no fee/charge for attending 
 IEEE technical seminar. You don&#39;t have to be an IEEE Member to attend. Ref
 reshments are free for all attendees. Please invite your friends and colle
 agues to take advantage of this talk.&lt;/span&gt;&lt;/p&gt;
END:VEVENT
END:VCALENDAR

